AN ABSTRACT OF THE DISSERTATION OF
Timothy L. Deboodt for the degree Doctor of Philosophy in Rangeland Ecology and
Management presented on May 8, 2008.
Title: Watershed Response to Western Juniper Control
Abstract approved by: _______________________________________________
John C. Buckhouse
Western juniper (Juniperus occidentalis) encroachment has been associated with
increased soil loss and reduced infiltration resulting in the loss of native herbaceous plant
communities and the bird and animal species that rely on them. Hydrologically,
however, change in water yield has been linked with the amount of annual precipitation a
site received. Studies published in the 1970’s and 1980’s, suggest that a minimum 4500
mm (18 inches) of annual precipitation was necessary before an increase in water yield
manifested itself following vegetation manipulation. In 1993, a paired watershed study
was initiated in the Camp Creek drainage, a tributary of the Crooked River of central
Oregon, to evaluate the impacts of cutting western juniper on the hydrologic function of
those sites. The study involved a paired watershed approach using watersheds of
approximately 110 hectares (270 acres) each to evaluate changes in a system’s water
budget following the reduction of western juniper. The 30 year average annual
precipitation for the area is 3500 mm (13.75) and during the study period, annual
precipitation ranged from 80 percent to 129 percent of average.
In 2005, following 12 years of pretreatment monitoring in the 2 watersheds (Mays
and Jensen) all post-European aged juniper (juniper < 140 years of age) were cut from the
treatment watershed (Mays). Analysis indicated that juniper reduction significantly
increased late season spring flow by 225 percent (alpha > .05), increased days of recorded
ground water by an average of 41 days (alpha > .05) and increased the relative
availability of late season soil moisture at soil depths of .76 m (27 inches) (alpha > 0.1).
Ephemeral channel flow did not show a predictable trend during 2 years of post
treatment measurements. Channel flow is dependent on spring snow melt and severe
summer thunderstorm activity. When winter soils were greater than 0 degrees Celsius
(32 degrees F), the source of channel flow in Mays was observed to be seepage from the
channel banks. Channel flow in Jensen appeared to be a result of rock forcing subsurface
flows to the surface.
Vegetative responses showed significant increases in perennial forb canopy cover
(alpha > .01) and annual forb and annual grass basal cover (alpha > .05). Increases were
also found in reduction of percent bare ground and increase in shrub cover, but were not
significant. A statistically insignificant decrease in perennial grass cover was noted in the
treated watershed however a large amount of reproductive culms were noted in the
treated watershed in 2007 compared to the control watershed.
Hillslope erosion and channel morphology showed no predictable trend following
treatment. Inherent differences in channel morphology between the two watershed prior
to treatment existed. This difference may be a product of the two channels being at
different evolutionary or successional stages relative to each other and thus indicating
that channel recovery would be different for each watershed.
The Camp Creek project illustrated that for this system, managing vegetation for
water yield may be obtainable at a much lower precipitation threshold than what was
previously reported in the literature.
©Copyright by Timothy L. Deboodt
May 8, 2008
All Rights Reserved
Watershed Response to Western Juniper Control
by
Timothy L. Deboodt
A DISSERTATION
Submitted to
Oregon State University
in partial fulfillment of
the requirements for the
degree of
Doctor of Philosophy
Presented
Commencement
May 8, 2008
June 2008
Doctor of Philosophy dissertation of Timothy L. Deboodt presented on May 8, 2008.
APPROVED:
Major Professor, representing Rangeland Ecology and Management
Head of the Department of Rangeland Ecology and Management
Dean of the Graduate School
I understand that my dissertation will become part of the permanent collection of Oregon
State University libraries. My signature below authorizes release of my dissertation to
any reader upon request.
Timothy L. Deboodt, Author
ACKNOWLEDGEMENTS
This research project was initially supported through a grant from the EPA with
Dr. John Buckhouse, OSU; Mel George, UC-Davis; and Sherm Swanson, UN-Reno
providing grant leadership. Additional funding was provided by Bureau of Land
Management Science Grant (thanks to Michelle McSwain, Prineville District) and the
Oregon Watershed Enhancement Board (OWEB). Other financial support was provided
by Crook County Taylor Grazing Board, Crook County Extension Service District and
Secure Rural Schools and Community Self-Determination Act of 2000: Title II.
I would like to first like to thank my family. In 2002, when I became involved in
this project, my oldest son Matthew was a senior in High School and Russell was a
sophomore and Grace Renee was in Middle School. Taking time away from their
activities was never an easy choice. Their support and understanding for all the missed
things is greatly appreciated. Thanks to them.
To my wife, Grace, thanks for all the support and encouragement you provided.
Words are impossible to find that adequately express my gratitude. While family was to
always come first, you were supportive when it could not. Thanks for the gentle pushes,
the constant love and understanding. Encouraging this journey could not have been easy
and it has been always appreciated. Thanks to you.
I would like to say thanks to the Crook County Extension office for all of their
support and understanding. This has been a 6 year journey. Many times, I am sure I
forgot to do things and you covered for me and never let on what you did. I appreciate
your interest in the project.
To Debby Maddy and Teresa Hogue, my supervisors with OSU Extension, your
support has not gone unnoticed. Debby you provided the encouragement and support to
allow this to happen. It is a wonderful thing to work for an organization that supports its
employees by encouraging them to be their best and to continue to develop
professionally. Teresa, over the last couple of years, your interest in the project and this
process did not go unnoticed. Thank you both.
The all the cooperators associated with this project, thank you, the list has grown
daily and it is impossible to thank each on individually. Special thanks to John Swanson,
Prineville District – BLM. John, you are perhaps the one most responsible for the
success of this project. Making sure we followed all the rules, allowing us to work within
the rules to accomplish some amazing things. Doc, Connie, Travis and Cynthia Hatfield,
I know that you were beginning to wonder if we would ever cut a tree. Your cooperation
has been wonderful, staying with us for all these years, supporting us, providing us with
carefully thought-out questions and occasionally reminding us to close a gate (ha-ha).
Earl McKinney and Butch (John) Heilmeyer, thanks for the inspiration and get it done
attitude. I still can’t believe how fast you cut all those junipers. Lynne and John Breese,
your insight into the practical applications of juniper harvest, editing and mentorship
have been greatly appreciated as well. To the McCormack family, Bill and Donna, Bill
and Jere, Jeff and Nin your friendship and support has helped to make this process a truly
positive journey.
To my committee, Dr. Mel George, Dr. Dave Thomas, Dr. Herbert Huddleston
and Dr. Steve Tesch, thank you for your continued guidance, support and critical review
of this project. You understood the uniqueness of this study, the conflicting schedules of
family, full time employment and the role of being a student and provided me with the
tools to balance it all. A special thanks to Dr. Fred Obermiller. Dr. Obermiller started
this project with us, bringing a level of excitement to the project, and understanding well
before I did the implications of its findings. Dr. Obermiller passed away long before he
could see its outcomes, but his memory will be here forever.
Mike Fisher, my mentor, my associate, my friend. Thank you for providing
leadership to this project from its beginning. Thank you for staying with the monitoring,
even when no one was noticing, even when no one was providing any financial
assistance. Your commitment has resulted a long term study, the only one of its kind in
kind in Oregon, contributing science based information to practical land management
applications. I am honored to be associated with you and this project.
To the Rangeland Ecology and Management Department, your support and words
of encouragement have been wonderful. Dr. Krueger, thanks for helping make all this
possible. Dr. Borman, thanks for taking the time to just sit and talk when necessary. Dr.
Lee Eddleman, thanks for being an inspiration.
Finally, I would like to extend the most heartfelt thank you to John Buckhouse.
Since the first time we met in 1978, you have been an advisor, mentor, peer and friend.
You are the reason I became involved with this project, you are the reason it has come to
a successful conclusion. I am forever grateful that you provided this opportunity to
explore and learn. I have learned much from you; it is now my hope that I can share that
knowledge with others. Thank you!
TABLE OF CONTENTS
Page
Introduction …………………………………………………………
Project Summary …………………………………………....
Project Description ………………………………………….
Project history ……………………………………….
1
1
2
3
Project Site Description……………………………………..
Location……………………………………………..
Soils…………………………………………………
Vegetation……………………………………………
Geomorphologic characteristics……………………..
Channel profiles and stream order…………………..
5
5
6
8
10
11
Project Objectives……………………………………………………
12
Literature Review……………………………………………………
Overview of vegetation manipulation for water yield………
Use of paired watershed approach…………………………..
Hydrologic function and western juniper……………………
14
14
20
22
Methods……………………………………………………………..
Reconnaissance……………………………………………...
Soil classification……………………………………………
Vegetation……………………………………………………
Precipitation…………………………………………………
Hydrology……………………………………………………
Channel cross-section ……………………………………….
Hillslope erosion process……………………………………
Geomorphometry……………………………………………
Additional monitoring added since 2003……………………
Ground Water Wells.………………………………………..
Spring flow………………………………………………….
Weather stations……………………………………………..
Soil moisture and soil temperature………………………….
Remote access of data……………………………………….
Location of monitoring sites…………………………………
Treatment……………………………………………………
28
28
29
30
31
31
31
32
33
33
34
35
37
38
40
42
43
TABLE OF CONTENTS (continued)
Page
Results ……………………………………………………………….
Weather………………………………………………………
Precipitation………………………………………….
Air temperature………………………………………
Wind…………………………………………………
Relative humidity……………………………………
Vegetation……………………………………………………
Soil moisture and soil temperature…………………………..
Ground water………………………………………………...
Spring flow…………………………………………………..
Channel flow…………………………………………………
45
45
45
47
47
48
50
51
64
68
70
Discussion……………………………………………………………
Weather………………………………………………………
Precipitation………………………………………….
Air temperature………………………………………
Wind…………………………………………………
Relative humidity……………………………………
Vegetation……………………………………………………
Soil moisture and soil temperature…………………………..
Ground water………………………………………………...
Spring flow…………………………………………………..
Channel flow…………………………………………………
Future research thoughts……………………………………..
73
73
73
75
76
77
77
79
82
85
87
88
Conclusions…………………………………………………………..
Soil moisture…………………………………………………
Ground water………………………………………………...
Spring flow ………………………………………………….
Channel flow ……………………………………………….
Vegetation……………………………………………………
Hillslope erosion……………………………………………..
Channel morphology…………………………………………
92
93
94
94
94
95
96
96
Literature Cited………………………………………………………
99
Appendices…………………………………………………………..
103
LIST OF TABLES
Table
Page
1. Federal versus Private Ownership of the Study Area……………………….
4
2. Soil series classification by percentage of total area in each watershed
(Fisher, 2004, pg 42)………………………………………………………...
6
3. Map Units for NRCS Soil Survey, 2007 (Draft)……………………………
8
4. Soil Mapping Units by Percent of Area……………………………………..
8
5. Significant P-values from ANOVA for variables describing the
differences in cover (%) between watersheds (Fisher, 2004, pg 43)………..
10
6. Aspect distribution classified by percentage of total area
Jensen and Mays watersheds (Fisher, 2004, pg. 64)…………………………
10
7. Frequency of stream order by watershed (Fisher 2004, pg.73)……………...
12
8. Annual water year precipitation for Mays and Barnes Station………………
46
9. Minimum and maximum temperatures for Mays and Jensen
Watersheds 2005-2007………………………………………………………
47
10. Comparison of the differences in means between Mays and Jensen
Watersheds for vegetative cover before and after treatment………………..
50
11. Comparison of before and after treatment means of the vegetative cover
values for Mays and Jensen watersheds for the NE and SW aspects……….
52
12. Mean soil moisture readings for the end of the year by probe location
for Mays and Jensen watersheds……………………………………………
58
13. A comparison of soil moisture mean differences between
Mays and Jensen for each year………………………………………………
58
14. A comparison of soil moisture mean differences between each of the
treatment years and the pre-treatment year..………………………………..
59
15. A comparison of the average of soil moisture mean differences for the
two post-treatment years and the pre-treatment year………………………..
60
LIST OF TABLES (Continued)
Table
Page
16. Number of days from September 1 in which the lowest soil moisture
Reading was recorded……………………………………………………….
62
17. The average number of days (from September 1) until soil moisture
began to accumulate at the various probe depths for each of the watersheds.
63
18. A comparison of the average number of days between Mays and Jensen
Watersheds when soil moisture began to accumulate for each year……….
63
19. A comparison of the difference in the average number of days before
soil moisture accumulation began by profile depth for each of the treatment
years and the pre-treatment year for Mays and Jensen……………………....
64
20. A comparison of the difference of the average number of days before
soil moisture accumulation for the combination of post-treatment years
with the pre-treatment year………………………………………………….
64
21. A comparison of the average number of days water was recorded in
each well. Pre and post-treatment years consist of 2 years each……………
65
22. Test for treatment differences in days of recorded water……………………
66
23. T-test for spring flow data, lowest recorded flow (gpm)……………………
69
LIST OF FIGURES
Figure
Page
1. Aerial photograph of project area, 2004…………………………………
3
2. Ownership of Project Area………………………………………………
4
3. Soil information, NRCS, 2006 Soil Survey (draft document)…………..
7
4. General relationship of vegetation to potential evapotranspiration (PET)
and precipitation (P) in the Colorado River Basin. The ranges in PET
and P are broader than indicated by the average values plotted for each
vegetation type. The dashed lines represent approximate marginal
conditions for improving water yield by manipulation of vegetation;
the potential is greater in the direction of warm and wet climate.
(Hibbert 1979 pg. 5)…………………………………………………….
16
5. Hydrologic budget components of pinyon and juniper watersheds from
estimates or complication of various studies
(Roundy and Vernon 1999 pg. 173)…………………………………….
23
6. Relative sizes of components of the forest hydrologic cycles, before
and after timber cutting (Fitzgerald 1997, pg. 59)………………………
24
7. Spring flow monitoring site, Mays with VORTFLOW/FLOW COMP©
36
8. Jensen weather station and flume……………………………………….
38
9. Installation of soil moisture probes……………………………………..
39
10. Location of monitoring stations………………………………………..
42
11. Types of treatments in Mays watershed in 2005……………………….
44
12. Camp Creek watershed Mays precipitation……………………………
45
13. Camp Creek watershed Jensen precipitation…………………………..
46
14. Jensen Weather Station maximum wind speed (miles per hour)……….
48
15. Mays Weather Station maximum wind speed (miles per hour)………..
48
16. Jensen Weather Station minimum relative humidity (percent)………...
49
LIST OF FIGURES (Continued)
Figure
Page
17. Mays Weather Station minimum relative humidity (percent)………….
49
18. Jensen Watershed soil moisture, end of the season probe readings
for the bottom profile….………………………………………………..
53
19. Mays Watershed soil moisture, end of the season probe readings
for the bottom profile…….……………………………………………..
53
20. Jensen Watershed soil moisture, end of the season probe readings
for the middle profile……………………………………………………
54
21. Mays Watershed soil moisture, end of the season probe readings
for the middle profile……………………………………………………
54
22. Jensen Watershed soil moisture, end of the season probe readings
for the top profile..……………………………………………………..
55
23. Mays Watershed soil moisture, end of the season probe readings
for the top profile…………………………………………….…………
55
24. Average end of season soil moisture reading for Jensen and Mays
watersheds, bottom profile………………………………………………
56
25. Average end of season soil moisture reading for Jensen and Mays
watersheds, middle profile………………………………………………
57
26. Average end of season soil moisture reading for Jensen and Mays
watersheds, top profile…………………………………………………
57
27. Days of recorded water in wells for the water years 2003-04 through
2006-07………………………………………………………………...
65
28. Days of recorded well water pre and post-treatment and the
associated differences………………………………………………….
66
29. Depth to ground water – Jensen wells 2003 - 2007……………………
67
30. Depth to ground water – Mays wells 2003 – 2007……………………..
67
31. Mays and Jensen spring flow ……………………………………………
68
LIST OF FIGURES (Continued)
Figure
Page
32. Late season springflows for Mays and Jensen …..…………………….. .
69
33. Number of days of yearly channel flow (January through June)………..
70
34. Springtime channel by year (January to June) in cfs………….…………
71
35. Number of days of annual channel flow………………………..………..
71
36. Total annual flow in cubic feet per second………………………………
72
37. Average daily soil temperature, Mays upper monitoring site……………
81
38. Average daily soil temperature, Jensen upper monitoring site…………..
81
LIST OF APPENDICES
Appendix
Page
A. NRCS Soil Classifications, Draft 2008…………………………………..
104
B. Partial Watershed Plant Species List (Fisher 2004)……………………...
123
C. Story of June 28 – 29, 2006 thunderstorm in Mays and Jensen………….
125
D. Vegetative monitoring changes from 2005 – 2007 ………………………
128
E. Vegetative responses, pictures from 7/25/07 – 7/28/07………………….
134
F. Soil moisture probe locations…………………………………………….
140
1
WATERSHED RESPONSE TO WESTERN JUNIPER CONTROL
INTRODUCTION
PROJECT SUMMARY
According to U.S. Forest Service publication PNW-RB-249, The Western Juniper
Resource of Eastern Oregon, western juniper’s dominance on eastern Oregon rangelands
has increased significantly since 1934 (Azuma et al. 2005). Azuma et al., (2005)
estimated that land occupied by western juniper has increased from 1.5 million to 6.5
million acres since the 1930s. Implications of this increase include loss of native,
herbaceous plant communities and the bird and animal species that rely on them,
increased soil loss, and reduced water infiltration. Based on water use models for
individual trees, the U.S. Forest Service estimated that mature western juniper tree
densities, ranging from 9 to 35 trees per acre, are capable of utilizing all of the available
soil moisture on a given site in a 13 inch precipitation zone (Gedney et al. 1999). Soil
erosion rates from sites with higher than the natural range of variability for western
juniper cover were an order of magnitude greater than similar sites that are within the
natural range of cover (Buckhouse and Gaither 1982). Research has shown that junipers
did increase soil loss rates due to the associated decline in herbaceous ground cover and
elevated surface runoff (Buckhouse and Gaither 1982; Bates et al. 2005). The juniper
canopy intercepts rain and snow, keeping it from reaching the ground thus making it
unavailable for plant growth, stream flow, or groundwater recharge; and they consume
large amounts of soil moisture. Previous monitoring of juniper control projects has
focused on changes in vegetative composition and productivity (Bates et al. 2005). These
studies have usually not monitored the hydrologic impacts of western juniper control.
2
This project was unique in that it involved a paired study approach to monitoring
changes in a watershed’s water budget following western juniper control. The value of a
paired watershed study is that the impacts of the treatment can be compared to the
untreated watershed. This study was unique in that it is the only long-term study of
western juniper ecosystems of its kind in the Pacific Northwest. Because of the time and
expense in monitoring treatment responses at the watershed level, such watershed
comparison studies are rarely undertaken. Similar studies in different ecological and
climatic zones have been conducted in Wyoming, Utah, Colorado and Arizona (Sturges
1994, McCarthy and Dobrowolski 1999, Bosch and Hewlett 1982) but no paired
watershed studies have been implemented in western juniper ecosystems.
PROJECT DESCRIPTION
The Camp Creek Paired Watershed study was initiated in 1993 to study the
effects of western juniper removal on sediment yield, water yield and vegetative
conversion (Fisher 2004). Two watersheds, Mays and Jensen were identified in the
Camp Creek drainage, a tributary of the Crooked River, Deschutes River Basin. Mays
and Jensen were named for the original homesteaders in the area. The study area was
created with the primary intent of calibrating and monitoring two watersheds for a period
of time for the purpose of understanding the comparative relationships of vegetation,
geomorphic and hydrologic parameters prior to treatment. Pretreatment monitoring and
analysis occurred from 1994 through 2004 (Fisher 2004).
3
Project History:
This project was initiated as part of a 3 state effort funded by an Environmental
Protection Agency (EPA) grant to look at arid land hydrologic issues in Oregon, Nevada,
and California. In 1993, the Mays and Jensen watersheds were selected and monitoring
of various attributes commenced. Each watershed was delineated on the upper bounds by
its ridge-tops and the lower ends designated by the placement of a channel flume. Mays
watershed is approximately 113 hectares and Jensen is approximately 106 hectares.
Figure 1. Aerial photograph of project area, 2004
The watersheds are located on the west branch of Camp Creek. Fourteen to
twenty-five percent of each watershed was under private ownership and the remaining
4
part study area is public land under the management of Prineville District, Bureau of
Land Management (BLM) (Table 1).
Table 1. Federal versus Private Ownership of the Study Area
Mays
Jensen
75% BLM
86% BLM
25% Private
14% Private
The general orientation of both watersheds is to the north. Livestock grazing
occurred in the project area and was administered by BLM under the guidance of the
Brothers-La Pine Resource Management Plan (RMP)(Figure 2).
Figure 2. Ownership of Project Area
5
Pretreatment monitoring of the area occurred from 1993 to 2004. Monitoring
parameters were vegetation composition, hillslope soil movement, and channel
morphology and flow. Precipitation was collected onsite and weather data was compared
with Barnes Station, a USGS weather station located approximately 10 miles east of the
project area. Fisher (2004) analyzed the comparative similarities and differences between
the two watersheds. These comparisons provided the basis for analyzing post-treatment
effects. As a result of pre-treatment analysis, additional parameters were added to the
data collection protocol; the monitoring of relative soil moisture, spring flow, and the
sub-surface distance to ground water were added in 2003.
PROJECT SITE DESCRIPTION
Location
The project site is located approximately 65 miles southeast of Prineville, Oregon.
Mays and Jensen watersheds are tributaries to the west branch Camp Creek, a tributary of
the Crooked River, a sub-basin within the Deschutes River Basin. The study area is
located within Section 32 and 33, T18S, R20E and Section 5, T19S, R20E Willamette
meridian. The area is located at the southern end of the John Day Ecological Province
(Anderson et al. 1998). The project site varied in elevation from 1370 meters to 1524
meters and the 30 year annual precipitation (1971 – 2000) at Barnes Station was 3500
mm. Sixty percent of the precipitation occured from October through March with only
25 percent falling during the growing season of April – June (Oregon Climate Service).
Temperatures range from mean daily maximum of 30 degrees Celsius in August to mean
6
minimum low of -7 degrees Celsius in February, with extremes recorded of 39 degrees
Celsius and -34 degrees Celsius.
Soils
Fisher (2004) found there were no published soil surveys for the study area.
Fisher worked with area BLM soil scientists to classify soils in the study area. Three (3)
primary soil series, Westbutte, Madeline, and Simaton were defined. Several other soil
series were noted but found to be in relatively minor amounts. Westbutte very stony
loam and Madeline loam, the two primary soil types, were found to make up
approximately 70 - 74 percent of the study area (Fisher 2004). Simaton gravelly silt
accounted for the balance. The Westbutte series was derived from weathering lava rock
and was found on predominately north facing slopes of 3 – 30 percent. The Madeline
series has a rooting depth of 30 – 50 cm and a greater potential for runoff and erosion
when compared to Westbutte. Simaton (gravelly silt loam) is a highly erosive soil.
Fisher (2004) reported that Simaton series contained a higher degree of clay when
compared to the other series. The soils in Jensen contained a higher percentage of
Simaton soils series when compared to Mays. Madeline soil series was found to make up
a large portion of the upper end of Jensen (Table 2). Mays watershed contained a high
percentage of Westbutte soil series which was less prone to erosion.
Table 2. Soil series classification by percentage of total area in each watershed
(Fisher, 2004, pg 42)
Soil Series
Jensen
Mays
Westbutte
Simaton
Madeline
Embal
26
21
48
0
50
3
20
1
7
In a review of hillslope erosion processes (soil movement), Fisher (2004) reported
no significant difference between the two watersheds (P-value = 0.694). Fisher reported
a weak to moderate relationship (r2 = 0.4174) for the two watersheds. For every 1 cm of
hillslope movement that took place within the gully systems in Jensen, there was
approximately 0.5 cm of soil movement in the gully systems in Mays.
In 2006, NRCS conducted soil surveys in the area (Figure 3). Evaluation of the
draft publication for the current soil classification data showed that Map Unit 246
occupies a significant part of both watersheds (56.7 % in Mays and 60.7 % in Jensen).
Map units 272 and 248 were also found in both watersheds. Map units unique to only
one watershed were: Jensen – 273, P500 and 430, Mays – 089 and 245 (Table 3 and
Table 4).
Figure 3. Soil information, NRCS, 2006 Soil Survey (draft document)
8
Table 3. Map Units for NRCS Soil Survey, 2007 (Draft)
Map Units – both watersheds
246
272
248
Anatone-Ateron-Polkbutte complex, 30 to 65 percent north slopes
Anatone-Bocker complex, 2 to 20 percent slopes
Anatone-Ateron complex, 30 to 65 percent slopes
Map Units – Jensen
273
Anatone-Bocker complex, 20 to 50 percent south slopes
430
Gerow silt loam, drained, 1 to 5 percent slopes
P500 Canest-Anatone complex, 0 to 10 percent slopes
Map Units – Mays
89
245
Luckycreek loam, 1 to 15 percent slopes
Anaton-Ateron-Polkbutte complex, 12 to 30 percent north slopes
Table 4. Soil Mapping Units by Percent of Area
Map Unit
246
272
248
273
430
P500
89
245
Misc
Mays
56.7
16.9
15.1
0.0
0.0
0.0
2.6
8.4
0.3
Jensen
60.7
5.3
9.1
10.3
7.5
6.9
0.0
0.0
0.2
Vegetation
Anderson et al. (1998) described the vegetative types associated with the John
Day Ecological Province of which the study area was a part. The historic range of
vegetation by cover would have been approximately 50 percent forest (pine, fir and
9
mixed pine-fir forests) with approximately 40 percent being nonforested and primarily
sagebrush-grassland. Western juniper would have occupied less than 10 percent of the
province. Common shrubs that were found in the area included mountain big sagebrush
(Artemisia tridentata, spp vaseyana), green and gray rabbitbrush (Chrysothamnus
viscidiflorus and C. nauseosus), and bitterbrush (Purshia tridentata). Grasses included
Idaho fescue (Festuca idahoensis), bluebunch wheatgrass (Agropyron spicatum),
Sandberg bluegrass (Poa secunda), prairie junegrass (Koelaria cristata) and Indian
ricegrass (Oryzopsis hymenoides). Forbs were numerous and highly dependent on
winter moisture to express themselves (Appendix B).
Analysis of pretreatment vegetation in the study area showed no significant
difference in vegetation composition when compared to aspect (Fisher 2004). Except for
percent perennial grass cover, stratification by aspect did not indicate differences in
percent cover of the various plant types between north and south slopes. Perennial grass
cover was reported to be 16.4 percent on north slopes and only 8.1 percent on south
slopes (significant P-value at alpha = 0.01)(Fisher 2004). The 3 variables most important
in controlling erosion processes (percent perennial grass, bare soil and tree cover) were
not significantly different between the two watersheds (Table 5). Fisher (2004) also
reported that there were no significant differences in plant community changes between
the two watersheds in regard to perennial grass, forb and tree cover from data collected in
1995 and 2003.
10
Table 5. Significant P-values from ANOVA for variables describing the
differences in cover (%) between watersheds (Fisher, 2004, pg 43).
Perennial grass
Bare soil
Litter
Live shrub
Dead shrub
Tree
Forb
Jensen
14.8
22.9
25.6
4.9
5.5
21.4
2.4
Mays
10.6
23.9
30.1
5.7
2.1
26.9
2.4
P-Value
0.145
0.585
0.962
0.591
0.053*
0.748
1.000
* shows significant differences at alpha = 0.05
Fisher (2004) reported tree density as high as 743 trees/ha and 680 trees/ha in
Jensen and Mays respectively. Tree density was reported to be highest in valley bottoms
and on toe slopes.
Geomorphologic characteristics
The topographical details of Mays and Jensen were similar. Slopes were similar
between the two watersheds with Jensen’s average slope equal to 25 percent and Mays’
average slope equal to 24 percent (Fisher 2004). Table 6 shows the relationship of aspect
between the two watersheds.
Table 6. Aspect distribution classified by percentage of total area
Jensen and Mays watersheds (Fisher, 2004, pg. 64)
Aspect
Jensen
Mays
North
East
South
West
36
31
5
23
33
17
11
26
11
Mays had approximately 10 ha more of south facing slopes and Jensen had 13
more ha of east facing slopes (Fisher 2004). Both south and east facing slopes had
similar characteristics, having dryer and harsher growing characteristics.
Orientation of the two watersheds was similar. Utilizing statistical slope
orientation diagrams (SSO), Jensen’s orientation is northerly and Mays was oriented
north by northwest (Fisher 2004). These northerly orientations should lend the
watersheds towards longer periods of frigid soils and maintenance of snow pack during
spring melting periods than if their orientation were southerly (Fisher 2004).
Frozen
soils often limit infiltration of snow melt and increase risk of overland flow during spring
snowmelt.
Channel profiles and Stream order
Channel profiles provided information about the watershed’s ability to transport
sediments out of the drainage area (Fisher 2004). Both watersheds exhibited similar
drops in elevation per length of channel (115 m/km). Within Jensen, there were greater
lengths of channel with less relief, lending support to the findings that within the lower
portion of Jensen, channel deposition was greater than the lower portion of Mays.
Stream orders can be used to determine the complexity of a given watershed
(Fisher 2004). Considering stream order frequency by watershed, since there are a higher
number of overall channel segments in Mays, channel erosion potential should be higher
than in Jensen (Table 7). Fisher (2004) reported that an increase in lower order stream
channels resulted in a decrease in time that it took runoff to reach the lower elevation
segments of the watershed.
12
Table 7. Frequency of stream order by watershed (Fisher 2004, pg.73).
Stream Order
#1
#2
#3
#4
Jensen
Mays
47
14
2
1
54
9
4
1
PROJECT OBJECTIVES
The purpose of this study was to quantify the impact – on a watershed scale – of
juniper control on the availability of water (quantity and timing) for beneficial uses
(water quality, fisheries, irrigation, recreation, etc.) as defined by Oregon State Statute.
The study involved a paired watershed approach for evaluating changes in a system’s
water budget following western juniper control. Water budget was measured in terms of
inputs (precipitation) and outputs (soil moisture, runoff, groundwater recharge and
evapotranspiration). Watershed impacts included the water budget impacts plus changes
in vegetation composition and cover, and erosion rates.
Monitoring water yield following juniper control had previously not been done in
the western juniper vegetation type. The value of a paired watershed study was that the
impacts of the treatment could be compared to the untreated watershed. The treatment
was to control western juniper in one of the watersheds. Juniper control included the
cutting of all post-European-aged junipers (juniper less than 140 years of age).
13
Study objectives were the following:
•
Measure hydrologic changes following juniper removal on a watershed
scale;
•
Evaluate changes in timing, duration and quantity of water expressed in
channel flow, spring output, ground water and soil moisture;
•
Calculate changes in hill slope and channel morphology following juniper
control;
•
Quantify changes in plant community composition following juniper
control.
In addition to changes in site condition, the wood products industry began to
develop an interest and commercial market in western juniper. As part of the treatment
activities, a harvest system was evaluated for costs of extracting juniper boles for use in
log homes, dimensional wood, and fence post/fire wood. Analysis of harvest information
provided land managers with information that can be used in determining opportunities
for adding value and benefits to juniper control projects (Dodson and Deboodt 2007).
14
LITERATURE REVIEW
OVERVIEW OF VEGETATIVE MANIPULATION FOR WATER YIELD
The importance of managing landscapes for water and watershed function has
been recognized for over 100 years. Water is an extremely important natural resource in
the western United States (Shepperd et at. 1992). The Organic Act of 1897, as reported
by Megahan, et al (2000, pg. 177) defined water as one the primary purposes for national
forests: “No national forest shall be established, except to improve and protect the forest
within the boundaries, or for the purpose of securing favorable conditions of water flows,
and to furnish a continuous supply of timber for the use and necessities of citizens of the
United States”. Huff et al (2000) described the management of forest vegetation and its
relationship to water yield as a long and lively issue. Research studies have focused on
gauged watersheds representing a variety of geographic locations, forest types,
topography and climate (Megahan et al. 2000; Hibbert 1979). For forested watersheds,
manipulating vegetation to increase water yield has been the primary focus of these
studies. Study results reported the effects of forests and forest disturbances on water
yield, peak and flood flows, snow accumulation and melt, soil erosion and numerous
water quality parameters including turbidity, nutrient loading and temperature (Megahan
et al. 2000).
Watershed studies have also been conducted on rangelands. Ice and Stednick
(2004) provided a summary of forest watershed research which has its roots in the Wagon
Wheel Gap study, a 16 year study initiated in 1911. Wilcox (1994) summarized work
done in the 1960’s and 70’s. Due to highly variable results in lower precipitation areas,
rangeland research has shifted from measuring the effects of management treatments to
15
measuring processes on a variety of management scales in order to identify predictable
hydrologic responses (Roundy et al. 1999). Hibbert (1983) indicated that increasing
water for onsite and offsite uses can be a viable objective for management of certain
western rangelands. As with forested water yield projects, replacing vegetation that uses
significant amounts of water with vegetation that uses less for the purpose of increasing
soil water percolation to streams and ground water is highly desirable. Since wet winters
are followed by dry summers in the Pacific Northwest, the effect of modifying the
vegetative community for water yield is of significant concern (Keppeler 1998). Many
residents in Oregon rely on springs and shallow wells for year-round water supply.
Beneficial uses of increased water flow also include fish and other stream fauna as well
as terrestrial species that depend on intermittent and perennial streams.
In central Oregon, United States Geological Survey (2000) concluded that ground
water in the upper Deschutes River Basin is hydrologically connected to surface flows in
the lower Deschutes River. Oregon Department of Water Resources (2002) has
implemented the Upper Deschutes Ground Water Mitigation Plan, which requires a
surface water mitigation plan before it will issue a water rights permit to pump from new
ground water wells. The National Marine Fisheries Service, a department of the National
Oceanic and Atmospheric Administration is currently working with Portland General
Electric, Confederated Tribes of Warm Springs and the Oregon Department of Fish and
Wildlife to re-introduce steelhead into the upper Deschutes River Basin above Pelton
Dam. This project is part of the Relicensing of the Pelton Hydroelectric Project.
Steelhead will be released into the Deschutes and Crooked Rivers in fall, 2007.
Maintenance of appropriate stream flows will be essential for successful spawning and
16
rearing activities in these watersheds. Hibbert (1979, pg. 2) stated “Public acceptance of
water yield improvement practices will partly depend on how people view the need for
more water (basically an economic issue) and on how they perceive the impacts of water
improvement practices on the forest environment, including the less tangible resources
such as wildlife and scenic beauty”.
Much debate has focused on the benefits of treating western rangeland for water
yield (Baker 1984; Wilcox 1994). Hibbert (1983) summarized findings by stating that
arid and semiarid rangelands are poor water producers because the potential
evapotranspiration (PET) is greater than precipitation (P). Figure 4 provides the visual
concept of how PET and P are related and in what type of plant communities, vegetation
manipulation for water yield activities should be focused.
Figure 4. General relationship of vegetation to potential
evapotranspiration (PET) and precipitation (P) in the
Colorado River Basin. The ranges in PET and P are broader
than indicated by the average values plotted for each vegetation
type. The dashed lines represent approximate marginal
conditions for improving water yield by manipulation of
vegetation; the potential is greater in the direction of warm
and wet climate. (Hibbert 1979 pg. 5)
17
Hibbert (1979) has set the minimum threshold for water yield response from
vegetation manipulation at 4500 mm (18 inches) of precipitation. In cooler climates
where precipitation is concentrated in winter months, the minimum threshold would be at
4000 mm (16 inches). Hibbert (1983) found that less than 1 percent of western
rangelands could be managed for water yield. Little or no increase should be expected
through the removal of low-density brush and pinyon-juniper woodlands.
Baker (1984) reporting on the long term watersheds project, Beaver Creek,
Arizona, found that for 8 years following herbicide treatment to control pinyon-juniper
(Utah juniper, Juniperus osteosperma (Torr.) Little) a statistically significant increase in
streamflow was measured. Herbicide treatment involved spraying 114 ha of Utah juniper
with a foliage applied application of 2.8 kg acid equivalent of picloram and 5.6 kg acid
equivalent of 2,4-D [(2,4 – dichorophenoxy) acetic acid] as triisopropolamine salts in 94
liters of water per hectare, August, 1968 (Baker 1984). A streamflow increase of 157
percent during the post-herbicide period was measured. This was compared to 11 years
of pretreatment data (1958-1968). Herbicide treatment killed 83 percent of the Utah
junipers. Shrub live oak and pinyon pine were damaged but recovered after 2 to 3 years.
Annual precipitation during the study period was 4630 mm with a high of 6910 mm and
low of 2100 mm measured. Winter precipitation (October – April) accounted for 64
percent of the total. Soils were developed from volcanic parent material and were
approximately 110 cm deep. The predominant soil type was Springerville Very Stony
Clay with clay texture throughout. Within this soil, the clay type was montmorillonite,
forming cracks as much as 5 cm wide and 1 m deep when dry.
18
Most of the water yield from the Beaver Creek experimental watersheds was
winter flow. Stream discharge was recorded during the winter 15 of the 25 years.
Overall, 21 of 25 years produced streamflow. The majority of flow occurred as snow
melt runoff in March and April. Soil moisture studies reported (Johnsen 1980) as a part
of this project that in the treated area, the soil below 30 cm did not reach the permanent
wilting point (-15 bars of atmospheric pressure) for 7 years after treatment but that in the
control area (untreated) soils dried to the permanent wilting point each spring.
Hibbert (1983) summarized two additional control methods conducted on the
Beaver Creek experimental watersheds. On one watershed, Utah juniper was controlled
by cabling and burning of the slash and in the second watershed, trees were felled by
hand and slash was not burned. Streamflow was not significantly increased in these
watersheds but more soil moisture was available for grass and forb production which
increased on both treated watersheds.
Sturges (1994) reported on findings related to manipulating sagebrush for water
yield. The 23 year study used herbicide to control sagebrush for water yield. A paired
watershed approach was used, and both watersheds had perennial flow with water
discharge being measured for 8 years before treatment and 14 years after treatment.
Water yield increased for 11 years in the treated area and then returned to pretreatment
levels. Two-thirds of the increase came from snowmelt discharge and the remaining
increase was associated with groundwater discharge through the remainder of the year.
The magnitude or date of the peak snowmelt discharge was not changed but the length of
the snowmelt period increased 22 percent. Grass production doubled in the first 5 years
and was 1.5 greater in the ninth and tenth years. Sagebrush density returned to
19
pretreatment levels after 11 years. Soils had developed from sandstone and were part of
the Mollisol and Aridisol order (Sturges 1994). Soils were loams in the A and B horizon,
the B horizon extending more than 1 m deep. Precipitation averaged 5180 mm per year
with 75 percent of the total falling as snow. Control of sagebrush did not affect snow
deposition. Change of snow water content was not significant (51.7 cm before treatment
and 54.9 after treatment). Increase in water yield was reported at 1.08 cm. Sturges
concluded that sagebrush control practices for water yield would be expected where soil
depth exceeds 0.9 m and precipitation is sufficient to fully recharge soil water. Long
term control of sagebrush would be necessary for treatment effects to persist.
The pinyon-juniper ecosystem is the most extensive forest type in the Colorado
River Basin (Hibbert 1979). This ecosystem type receives 3000 to 4500 mm (12 to 18
inches) of annual precipitation. Summer rains account for half or more of the annual
precipitation in the lower basin; winter rain and snow provide the majority of
precipitation in the upper basin. The potential for increasing water yield in this type of
ecosystem is negligible although small increases of less than 12 mm may be possible on
the wettest sites.
Swank (1969) in a study of the potential to increase water yield in the Ochoco
Creek watershed (Prineville, Oregon) estimated that increased water yield varied from
.03 acre-feet/acre per year with juniper control to .8 acre-feet/acre per year with
lodgepole pine control. Ffolliott and Brooks (1988) noted that increases in annual water
yields generally occurred on watersheds characterized by the removal of tree basal areas
in excess of 100 ft2 per acre.
20
USE OF THE PAIRED WATERSHED APPROACH
A paired watershed study involves at least two watersheds, control and treatment
and two time periods – calibration and treatment. The purpose of the control watershed is
to account for seasonal and year to year weather variations as the management activities
remain the same. The treatment watershed has the changed management imposed on it at
some time. During the calibration phase, no change of management occurs in either
watershed. For a period of time, data are collected in both watersheds. In the treatment
phase, one watershed receives the change in management while the other watershed
remains with the original management. The strength of the study is in the amount of time
that base-line data (pre-treatment) are collected. As noted earlier, studies in Arizona and
Wyoming included 11 and 8 years respectively of pretreatment monitoring (Baker, 1984;
Sturges, 1994).
EPA proposes that there are 3 questions to answer before shifting from the
calibration phase to the treatment phase:
1) Is there significant relationship between the paired watersheds for all
parameters of interest?
2) Has the calibration period continued for a sufficient length of time?
3) Are the residual errors about the regression smaller than the expected
BMP effect?
The use of the paired watershed study approach is well documented in forest and
hydrology (Hewlett et al. 1969; Bosch and Hewlett 1982; Wilcox 1994; Monteith et al.
2006). EPA (1993) published guidelines for establishing paired watershed studies. The
first watershed experiment began in 1911 at Wagon Wheel Gap in Colorado. The paired
21
watershed approach utilized a control and treatment watershed design to quantify the
effects of deforestation on the volume and timing of streamflow, soil erosion, and
sediment loading (Megahan and Hornbeck 2000).
The strength of the paired study approach is perhaps also its short coming. Alila
et al (Web site accessed June, 2007) summarized the arguments against traditional paired
watershed studies:
1) Lack of control from not-so-trivial but often extremely important
differences between the physical characteristics of the control and
treatment watersheds.
2) Lack of control from a spatially and temporally varying climate.
3) Short length of pre- and post-treatment record that cause statistical
tests to be inconclusive.
Alila et al. (Web site accessed June, 2007) argued that the traditional method of
measuring precipitation inputs and stream outputs fails to explain the internal catchment
processes whether they be above or below ground. The “black box” approach makes
transferring knowledge gained from a paired watershed study to other, ungauged nearby
watersheds difficult. The authors of that paper suggested it is impossible to explain the
extra variability in the relation between percent forest removal and the magnitude of the
hydrologic impact if only precipitation/streamflow are measured. These authors also
suggested that the use of modeling that incorporates detailed field work done throughout
the watershed will provide a more enlightened look at the relationships of management
activities and response. Modeling provides the opportunity to reduce
22
calibration/treatment periods and allows the researcher to hold constant climate variables
and vegetation changes associated with time.
HYDROLOGIC FUNCTION AND WESTERN JUNIPER
The hydrology of a given watershed is a function of the precipitation amount,
intensity and seasonality, watershed geomorphic characteristics including topography and
soil characteristics, and vegetation composition and complexity (Bowns 1999). Bedell
and Borman (1997) credit Hugh Barrett (retired Rangeland Management Specialist,
Oregon BLM) for describing and coining a phrase germane to the essence of watershed
management as “the soil’s ability to Capture, Store and Safely Release” precipitation
which is delivered to a site. Watershed management extends beyond the concept of
water management to include the importance of other resources and how they are
interconnected. Those resources include plant, animal, soil and climate (expressed
through precipitation, temperature, wind, solar radiation, etc.). Capture refers to the
processes in which precipitation enters the soil profile. Storage focuses on the moisture
within the soil profile. Safe release is the process of either soil moisture being used for
plant growth or excess soil moisture moving through the soil profile, recharging ground
water and expressing itself as either flow from springs or as contributing to streamflow
through seepage from stream banks (Bedell and Borman 1997).
Vegetation plays a critical role in connecting precipitation with the soil profile.
Vegetation serves as a buffer, reducing raindrop impact on the soil surface, adding to the
soil surface roughness (litter and plant biomass) to slow overland flow and increase the
amount of soil/water contact time, thus increasing opportunities for infiltration and by
23
increasing soil water holding capacity through the addition of organic matter into the soil
profile. Land managers, landowners and public land management agencies, through the
application of management plans, impact vegetation’s ability to buffer precipitation
inputs and aid in the watershed’s ability to capture, store and safely release water.
As discussed earlier, forest hydrologists have illustrated the benefit of forest tree
reduction for the purposes of increasing water yield (Bosch and Hewlett 1982; Hibbert
1983; Wilcox 1994). While Baker (1984) illustrated benefits of removal of juniper in
Arizona, there is general consensus that little water benefit will be realized in much of the
intermountain west where annual precipitation varies from 2000 mm to 5000 mm
(Hibbert 1979; Roundy and Vernon 1999). However, relatively small amounts of water
yield can be critically important to sustain both livestock and wildlife during the late
summer months. Based on various studies, Roundy and Vernon (1999) provided a
conceptual water budget for these systems (Figure 5). Fitzgerald (1997) provided a water
budget for forested systems before and after tree harvest (Figure 6).
Figure 5. Hydrologic budget components of pinyon and juniper
watersheds from estimates or complication of various studies (Roundy and
Vernon 1999 pg. 173).
24
Figure 6. Relative sizes of components of the forest
hydrologic cycles, before and after timber cutting
(Fitzgerald 1997, pg. 59).
The impacts of expanding western juniper woodlands or juniper control
treatments on hydrology and water balance are not well understood (Miller et al. 2005).
Gedney et al (1999) found that landowners and land managers have reported small
streams and springs drying up or have stopped flowing as juniper became established on
their rangelands. It has been suggested that due to the interception of precipitation and
transpiration by juniper, decreased flows are the result of reduced recharge of the shallow
aquifers that feed the ephemeral streams and springs (Gedney et al. 1999; McCarthy III
and Dobrowolski 1999; Larsen 1993). Roundy and Vernon (1999) suggested that
although sites occupied by juniper tend to be areas of low precipitation and high
evapotranspiration, there may be sites where a shallow impermeable layer may allow for
seasonal saturation and sufficient interflow to feed local springs and streams. These and
other similar, anecdotal stories provide the current basis for discussions regarding impact
25
of western juniper on a site’s hydrologic function and the subsequent site response
following juniper removal.
Western juniper has increased significantly in the last one hundred years in both
the area and density that it now occupies. Azuma (2005) reported that western juniper
has increased in eastern Oregon from 600,000 hectares to over 2.4 million hectares (1.5
million to over 6 million acres) since 1935. The increase in acreage is attributed to many
factors but focus primarily on; 1) periods of higher precipitation and cooler temperatures
in the late 1800’s that favored seed production, germination and survival of juvenile
trees, 2) uncontrolled, excessive grazing by domestic livestock in the late 1800’s through
the early 1900’s which removed grasses and favored shrub production (a primary host
plant for juniper seedlings), and 3) reduction of fire as a result of grazing removing the
fine fuels and humans increasing fire suppression efforts since the early 1900’s (Miller et
al, 2005).
As illustrated in Figure 5, interception is the first barrier to precipitation entering a
site. Canopy interception is the portion of gross precipitation that is retained by the
canopy and eventually lost by evaporation. Larsen (1993) provided a summary of studies
conducted on various juniper species and storm intensities throughout the intermountain
west. He reported that interception rates varied from 4 to 70 percent with the average
being 14 percent. Average annual precipitation lost to interception was estimated to be
17 percent across many studies.
Larsen (1993) also looked at the role of litter as an effective interceptor of
precipitation. Beneath the junipers litter depths varied from 10 to 60 mm (canopy edge to
near tree bole). Of the measured throughfall and stem flow of moisture measured coming
26
from below the tree canopy, average water penetration into the litter layer was 10 to 31
mm (measured from canopy edge to near tree bole). Larsen suggested that interception
of precipitation from the litter may be higher than losses from the tree itself. Larsen
found that this condition created small islands (beneath individual trees) where
precipitation inputs rarely reached mineral soil in the juniper woodland. The
combination of tree interception plus litter (beneath the tree) interception would support
the conclusion that effective or net precipitation is a reduction of 1:1 relationship of with
canopy cover and gross precipitation. Eddleman et al. (1994) stated that small storms are
likely to produce no stem flow and low throughfall which may be held in the litter. Net
interception with subsequent evaporation or sublimation back into the atmosphere will be
a high percentage of input for small storms.
Infiltration studies are limited (Eddleman et al. 1994). Infiltration rates in juniper
ecosystems are comparable and not significantly different than sagebrush or grassland
ecosystems. Mean infiltration values for these types of systems vary from 44 mm/hr to
66 mm/hr. Wood et al (1987) found a positive relationship between infiltration rates and
grass production, litter, total ground cover and soil sand fraction near the soil surface and
a negative relationship with soil moisture, bulk density and silt fraction near the soil
surface. These results support the premises of Bedell and Borman (1997) that vegetation
can play a significant role in controlling watershed function. Others also report the
positive attributes of vegetation in affecting infiltration. Eddleman et al. (1994) found
multiple studies which indicated that factors which shorten slope and increase detention
storage (such as miniature terraces and plant density) as well as other organic or
inorganic objects on the soil can cause infiltration rates to increase. Minimizing slope or
27
organic debris dams hold water on the soil surface longer, thus increasing soil/water
contact times and allowing for more water to move into the soil profile.
Opposite of infiltration is overland flow. Excessive overland flow would not be
considered desirable within the concept of watershed function and the safe release of
water. With overland flow, the concern is the potential to carry sediment and transport it
off site. Sediment transported off site removes limited nutrients from the site and reduces
site productivity (Eddleman et al. 1994). Buckhouse and Gaither (1982) illustrated the
potential soil loss risk as it related to storm size and vegetative community type. Soil
erosion rates from sites with higher than the natural range of variation for western juniper
cover was a magnitude of order greater than similar sites that were still within their
natural range of variation of cover.
Western junipers are evergreen and thus have the potential to utilize soil moisture
throughout the year. Juniper maintains its ability to photosynthesize and transpire water
whenever weather conditions are right and provides it a competitive advantage over
grasses, forbs and some shrubs that must annually produce new leaves before
photosynthesis can begin. Jeppesen (1977) found that if subsoils are not frozen, western
juniper can use soil moisture throughout the year. Winter soil moisture use by juniper is
from deep soil moisture, greater than 0.51 m (Jeppesen 1977). Soil temperatures at this
depth remain relatively constant 1 to 4 degrees Celsius. This winter moisture use
decreases herbaceous plant production during the spring and summer months. Miller
(1990) reported that as long as the soil and air temperature was above freezing and soil
moisture was available, western juniper was able to utilize soil moisture while other plant
species were dormant.
28
METHODS
The Camp Creek Watershed Study was established in 1993 to quantify and
understand how vegetative conversion (removal of western juniper) could affect
hydrologic functions (Fisher 2004). Monitoring protocols were established in 1993 for
the purpose of collecting baseline data for which the subsequent treatment phase of the
study could be compared. Fisher (2004) provided the detailed explanation of those
monitoring protocols. Protocols for reconnaissance, soil classification, vegetation,
precipitation, hydrology, channel cross-sections, hillslope erosion processes and
geomorphology are summarized here.
The calibration phase of this study provided the ecological and geomorphologic
characteristics of the two watersheds. Protocols were developed to provide information
about vegetation, soils, topography, geology, channel morphology streamflow, local
climate and erosion processes (Fisher 2004).
RECONNAISSANCE
A general survey of the two watersheds provided initial information which was
useful in developing initial sampling methods. Each watershed was stratified by slope,
elevation and aspect. Geographical Information System (GIS) was used to provide
assistance in interpretation of topographic maps and aerial photos. Boundaries of the
watersheds were recorded with the assistance of Global Positioning Systems (GPS) along
with the permanent locations of channel cross-sections, hillslope erosion plots and
29
vegetative transects. Significant land features (streams, springs) and man made features
such as roads were also mapped. Aerial photographs along with GIS interpretive
assistance provided information useful in establishing vegetation plots and monitoring
protocols. Ground truthing was used to validate and modify interpretations made from
photos and GIS.
Aerial photography was used to provide information regarding the extent of
western juniper (canopy cover). Topographical information was categorized into three
subjects; aspect, slope and elevation. Aspect was classified into 5 units; north, south,
east, west and level. Slope (rise over run) was classified into 6 units; 0-20 %, 21-40 %,
41 – 60%, 61-80 %, 80 – 100 % and 100 %+.
SOIL CLASSIFICATION
Original soil associations were described in conjunction with BLM soil scientist
Larry Thomas (1995). At the initiation of the study, there was no soil survey information
available. The original USDA, Crook County Soil Survey (1966) only incorporated soil
information for the irrigated lands in the western one-third (Prineville area) of the county.
The soils in the study area originally were mapped using ground reconnaissance and
aerial photographs. Information was transferred into electronic data bases for analysis by
GIS (Fisher 2004). In 2006, the soil survey team of the Natural Resources Conservation
Service mapped the area and provided a draft report in 2007 (NRCS Crook County Soil
Survey 2007 draft).
30
VEGETATION
Vegetation measurements were obtained utilizing the line-intercept method.
Transects were permanently marked with angle iron and steel fence posts and GPS.
Transects were stratified by slope and aspect (north, south, east and west). Transects (30
m) were established perpendicular to slope. Data from transects were collected 3 times
prior to treatment (1995, 2003 and 2005) and once post treatment (2007). Basal cover
was measured for perennial and annual grasses and foliar cover was measured for forbs,
shrubs and trees. Bare ground was measured. Litter including moss, needles and woody
material greater than 2 cm was recorded. Rock was measured if rock was equal to or
greater than 13 cm at point of intercept. Species composition was estimated using ocular
reconnaissance and western juniper frequency and stand density measurements were
obtained using aerial photographs and GIS interpretation. In 2005, juniper densities in
each watershed were measured through 160, 1/10 acre plots evenly blocked by aspect and
cover (high and low). Within each block, 5 plots were measured for detailed tree
information was collected including tree height, canopy width, diameter at breast height
and designated as either sapling, juvenile, mature (pointed and rounded) or decadent.
PRECIPITATION
Precipitation was measured in each watershed utilizing a Belfort Instrument
Company “Universal Rain Gauges”. Gauges to measure precipitation volume, storm
intensity, and frequency were placed near each flume. Data were collected to an
accuracy of 1 mm at 6-hr interval over 30-day periods. Precipitation data also were
31
collected from USGS weather station “Barnes Station” through the Oregon Climate
Service, Oregon State University.
HYDROLOGY
A 3.0 H-Flume was placed at the “bottom” or lower end of each study area for the
purpose of recording flow leaving the study area. The actual location of each flume was
controlled by channel morphology and site accessibility. A fabricated steel approach was
attached upstream of each flume for the purpose of stabilizing flow. Data were collected
utilizing a Druck© pressure transducer, placed in a stilling well which was connected to
the flume. The purpose of the stilling well was to buffer the measuring device. Flow
measurements were recorded utilizing Campbell Scientific CR10 data loggers. Flow
measurements were set to collect temporary data every 10 seconds. After every 10
minute interval, minimum, maximum and average stage were sent to the data logger for
storage. A factor of .05 inch or 1.27 mm stage height was established as the minimum
number (not zero) to record as “flow”. This protocol limited the collection of “zero” data
which can fill a database very quickly. Time of day, Julian date, and year were recorded
along with 24 hour average, minimum and maximum stage measurement. Conversion
from stage height to flow volume was calculated using conversion tables provided by the
flume manufacturer.
CHANNEL CROSS-SECTIONS
To measure changes in channel morphology, 25 channel cross-sections were
placed in the primary channels of each watershed upstream from the flume. After
32
initially identifying the first cross-section through random selection, cross-sections were
placed at 30 m intervals. Permanent stakes marked the location of each cross-section. A
stationary steel tube (25mm2) was used to measure the relative width and depth of the
channel at 100 mm intervals. With the use of a carpenter’s level to assure that the steel
tube and steel angle iron were level and perpendicular to each other, measurement from
ground to top of steel tube was recorded to nearest cm. At time of channel cross-section
establishment, photos upstream and downstream were taken. Channel gradient between
each cross-section was measured using a clinometer. Measurements were taken twice a
year, after snow melt and spring channel flow had ceased, and then again after the
summer thunderstorm period was over.
HILLSLOP EROSION PROCESSES
Sedimentation plots were used to measure annual or semi-annual active, subwatershed level erosion processes. Sampling was established in sub-watersheds that had
existing rills or small gullies and exhibited evidence of surface flow from the hillslope to
the main channel at the time of the study’s initiation. Twelve sub-watersheds were
identified in each of the main watersheds. Three (3) sedimentation rods were placed in
each sub-basin utilizing a systematic, randomized approach. Within the first 0 - 10 m of
the sub-basin’s intersection with the main channel, the first rod was randomly placed.
The second and third rods were then placed at 20 m intervals. For this study,
sedimentation was defined as the deposition or erosion of soil. Measurements were
recorded for both the upslope and downslope side of the rod. Since 2000, measurements
were taken after snow melt and channel flow had ceased and then again after the summer
33
thunderstorm period was over. Prior to 2000, measurements were taken once a year.
Some points were replaced due to loss of the stake (for unknown reasons) or stake was
destroyed by someone or something bending it over.
GEOMORPHOMETRY
Website data were accessed for the purpose of obtaining Digital Elevation Models
(DEM). The study area was located within the United States Geological Service (USGS)
Logan Butte Quadrangle. The DEM for this quadrangle was obtained and imported into
MicroDem© software. This software along with the DEM provided the opportunity to
delineate watershed boundaries, measure stream channel lengths, and analyze
geomorphologic characteristics.
ADDITIONAL MONITORING ADDED SINCE 2003
As the end of the pretreatment monitoring period was nearing, additional
monitoring/data needs were identified. Additional monitoring activities would measure
the additional ways that water conditions could change in the watersheds and would show
if those conditions changed following juniper control. Initially limited by money for
purchase and installation, additional grants obtained from the BLM and Oregon
Watershed Enhancement Board (OWEB) provided the opportunity to expand the
monitoring capabilities. Within each watershed, a spring had been identified but output
was never measured. In 1995, a series of shallow wells were placed at the bottom of each
watershed near the flume (5 wells per watershed). Well depth varied between 0.9 and 8.2
34
m. The collection of well data was sporadic and incomplete but free standing water in at
least one of the deeper wells in each watershed was recorded sometime during the year.
Soil moisture had been shown to increase and remain available longer for plant
growth following juniper removal (Hibbert 1983; Baker 1984). Jeppesen (1978) noted
that if soil temperatures didn’t freeze, western juniper was capable of utilizing soil
moisture throughout the winter period. With the exception of precipitation, weather data
was limited for the site. Winter snow accumulation had been shown to change with
vegetation type changes (Sturges 1994). Temperature and wind were important factors
in determining initiation of plant growth and evapotranspiration rates.
GROUND WATER WELLS
In the fall of 2003, USDA Forest Service personnel from the Ochoco National
Forest provided the expertise and drilling equipment to install 6 ground water wells per
watershed. A Simco© track portable drilling rig was equipped with a 5 inch auger drill
bit. Wells were drilled to a maximum potential depth of 8.2 m (27 ft.). Well depth was
less if bedrock was shallower or in the case of several holes in Mays, ash soils at about 6
m, collapsed the hole and filled in the bottom 0.5 - 0.9 m of the hole. Following the
drilling of each hole, washed, crushed rock (20 – 25 mm (3/4 – 1 in)) was place on the
bottom to a depth of approximately 50 mm. Each hole was cased with 50 mm (2 in.)
perforated PVC pipe. Pipe was perforated with the use of a circular saw, cutting slits into
the pipe every 0.15 m (6 in.). The pipe was rotated and slits cut on the opposite side of
the pipe as well. After the pipe was installed, washed and crushed rock (20 – 25 mm)
was placed on the outside of the pipe in the hole. The top of pipe was left above ground
35
approximately 0.30 to 0.45 m (1 – 1.5 ft.). The top of the pipe was capped with a screw
type cap to keep debris from falling into the pipe. Each pipe was marked with a 1.68 m
(5.5 ft.) steel t-post.
Beginning Jan. 2004, water depth was measured from the top of the pipe utilizing
a SoilTest© DR-760A water level indicator. This instrument was a portable, selfcontained transistorized instrument for determining water levels in bore holes up to 300
ft. When the probe made contact with water, an electrical circuit was established and the
ammeter indicator deflected from zero to full scale. The electrical cable was marked in
1.52 m (5 ft) sections and a tape measure was used to provide detailed measurements
between marks (US Army Corp Eng., 1980).
To allow for remote monitoring, a LEVEL-WATCH© sensor was placed in each
well on May, 2005. LEVEL-WATCH© sensors recorded the liquid level above the
sensor and was measured indirectly by recording the hydrostatic head of pressure
(Automata Service Manual, 2005). A vented tube was available for barometric pressure
compensation. Recording level of each sensor was 0 – 10.54 m.
SPRING FLOW
Throughout the pre-treatment period, two springs (one in each watershed) were
identified. Flow had never been recorded and in conversations with the person who is
both the landowner and BLM permittee, the springs had been identified as “seasonal at
best” (Hatfield, personal communication 2001). That is, each spring only flowed during
spring snowmelt periods and provided no flows from mid-summer through the winter.
In the fall of 2003, permission was received from the Prineville BLM District and the
36
private land owner to improve the two springs and install spring boxes and pipe so that
water could be collected and measured. A private contractor was hired to excavate and
install the appropriate materials. Upon completion of the installation, a seed mixture of
grass species was broadcast over the excavation sites to minimize erosion and weed
infestation. In Mays, two 38 mm (1.5 in) PVC pipes were plumbed into the spring box.
One pipe was placed approximately 0.15 m from the bottom of the spring box and the
other pipe was placed 0.40 m from the bottom of the spring box. Jensen spring had one
38 mm (1.5 in.) PVC pipe plumbed at 0.15 m from the bottom of the spring box. Flow
was periodically measured using a stop watch and 1.89 liter (2 qt.) pitcher. The amount
of time required to fill the pitcher was recorded. Values were converted to liters per
minute.
To allow for remote access of each spring, a VORTFLOW/FLOW COMP©
sensor was installed at the end of each outflow pipe (Figure 7). The sensor was
calibrated to read flows from 3 – 100 gpm. When flows dropped below 3 gpm, flow was
not detectable by the sensor.
Figure 7.
Spring flow monitoring site, Mays with VORTFLOW/FLOW COMP©
37
WEATHER STATIONS
With the exception of precipitation, the need for additional weather data was
identified. Temperature, wind speed and snow accumulation were important variables in
analyzing effects of the treatment. Two (2) weather stations were installed (one in each
watershed) near the location of the flume in the fall of 2003. Monitoring sensors
associated with each weather station were precipitation (tipping bucket), solar radiation,
relative humidity, leaf wetness, temperature, snow accumulation and wind speed / wind
direction. The following is a description of each of the sensors.
Sensor Name
Leaf Wetness
Wind Speed
Wind Direction
Rain: Weight
Rain:
Air Temperature
Relative Humidity
Solar Radiation
Snow Depth Sensor
Model
LEAFWET
200WS/WINDCOMP
200WS
1040-41
RW100
AD592
H50
LI200SZ
TS-15S
Range/Units
0-100 %
0-100 mph
0-360o
K/gm
.01 inch
Deg F
0-100%
0-2000W/Mˆ
0.006768 inch
Sensors were supplied by Automata Inc., Nevada City, California. Installation was done
by Wy-East RC&D technicians from The Dalles, Oregon. Sensors were mounted on a 50
mm diameter pole. The pole was 4.5 m tall and was built to pivot in the middle to allow
access to sensors for maintenance and calibration. Solar panels were used to keep a 12
volt deep cycle battery charged and functioning (Figure 8). The Belfort Instruments
Universal Range Gauges continued to be used and maintained as a redundant, back-up
system (Fisher 2004).
38
Figure 8. Jensen weather station and flume
SOIL MOISTURE AND SOIL TEMPERATURE
In semi-arid watersheds, soil moisture may be the only measurable hydrologic
response following vegetation conversion (Hibbert 1983; Wilcox 1994). Soil moisture
and soil temperature monitoring were added to the monitoring protocol in May, 2005.
Soil moisture sites were placed in two locations within each watershed. At each site, 3
soil moisture stations were located and at each station, 3 soil moisture probes were
installed. Soil probes were installed at 0.2, 0.45 and 0.76 m of soil depth. At each
station, a hole, 1m x 1.5m x 1.2m (width x length x depth) was dug utilizing a mini,
track-excavator. Probe holes were then drilled into the wall of the large hole utilizing a
16 mm drill bit (5/8 inch) (Figure 9).
39
Figure 9. Installation of soil moisture probes
The hole was just slightly smaller that the diameter of the probe (3/4 inch) so that the soil
to probe contact would be complete. This also assured that the soil around the probe was
disturbed as little as possible. According to the manufacturer, disturbing the minimum
amount of soil when installing the sensor reduces the stabilization time after installing the
sensor (Automata Service Manual 2005). Sensors were installed in May, 2005 as
recommended by the manufacturer.
The soil moisture probe used was an AQUA-TEL-TDR© probe. The AQUATEL-TDR© probe measured soil moisture reliably in all types of soils at all moisture
levels (Automata Service Manual, 2005). Overall probe length was 0.7 m in length
(sensor length is 0.48 m). The probe read the percent of available moisture by volume,
independent of soil texture. The AQUA-TEL-TDR© probe measured the dielectric
constant, which was directly related to water content in reasonable growing conditions
(Automata Service Manual, 2005). Readings obtained from the probe represented the
40
average dielectric constant, and were converted to percent moisture over the length of the
0.48 m active segment.
Sensors were calibrated in the factory utilizing a full emersion approach
(Automata, 2005). Probes were placed in water to assure reading of 100 percent. In the
field, probes were not calibrated. The data provided a picture of relative values, and are
useful in illustrating how soils dry down and rewet. At each soil moisture site, 2 of the 3
probes placed at the 0.2 m depth included a temperature sensor (AQUA-TEL-TDR+T).
The AQUA-TEL-TDR+T reported data in degrees F and soil moisture data from sensors
AQUA-TEL-TDR and AQUA-TEL-TDR+T were reported in percent moisture (0-100
percent).
REMOTE ACCESS OF DATA
The Paired Watershed Project was located approximately 65 miles southeast of
Prineville, Oregon. During the pre-treatment monitoring phase, it became evident that
taking regularly timed data was critical in understanding the hydrological processes that
were occurring at the site. Channel flow data were being recorded every 10 minutes but
there would be times that the data logger would quit working and data would be “lost”.
Traveling to the site daily or weekly was not possible so an effort was made to find a
system that was compatible with the sensors installed. Remote access provided a means
of accessing data and, if possible, provided a way to monitor the function of the
equipment so that it could be determined, in real time, if sensors were working or not,
which allowed for timely maintenance and repair.
41
Prior to choosing a system, several were reviewed and included cellular phone
connections, radio, combination of radio and cellular phone and satellite radio. Due to
the remoteness of the site and topography, there was no cellular signal near any of the
monitoring sites. Radio access was limited and required licensing and locating sites for
towers as well as getting through the permitting process. A combination of short distance
radio (monitoring site to ridge top) and cellular phone was also evaluated. Vegetation
(trees) limited signal quality and the cellular phone was limited to analog technology.
Cellular phone companies in central Oregon at the time were abandoning analog
technology in favor of digital. Working with Automata, Inc., satellite radios provided the
solution. The only requirement of this system was an unobstructed view of the sky so
that the radios could communicate with the orbiting satellites (Automata Service Manual,
2005). The satellite radio allowed access to all sensors daily, each on its own schedule.
Satellite radios require no repeaters and each radio was separate and transmitted data
through the satellite to a data server. Data were accessed through the internet web site:
http://ifpnet.com.
Nine (9) stations were located throughout the study area (4 in Mays and 5 in
Jensen) and served as focal points of all the sensors. Weather Stations were accessed
hourly, hydro stations (including wells, spring flow and channel flow) were accessed 4
times per day (every 6 hours), soil moisture sites were accessed 4 times per day (every 6
hours) and Jensen spring flow was recorded every 12 hours (twice per day).
Programming of each site allowed for increases or decreases of each station’s
accessibility but the relative value of the data compared to the cost was used to set the
42
schedule. Cost of operation was based on the number of times each sensor was accessed
per day.
LOCATION OF MONITORING SITES
Figure 10 shows the location of the monitoring stations in which the data was
collected through the satellite telemetry. The flume and well locations marked the lowest
end of each of the watersheds.
Figure 10. Location of monitoring stations
43
TREATMENT
In the fall of 2005, all post European aged juniper (trees less than 140 years old)
were cut in Mays. Cutting instructions reflected the current understanding and practice
associated with restoring pre-juniper dominant vegetative communities and watershed
function. In addition to the cutting of all post European aged juniper, all old growth
juniper was to remain on site. Old growth was defined as juniper with rounded tops,
deeply folded bark, and moss and lichens growing on the limbs. Additionally, any
juniper that was observed to be host to any wildlife was not cut.
The decision of which watershed to treat was made following extensive site visits
and interviews with geologists, rangeland managers and researchers. Surface evidence of
westerly sub-surface fracture orientation in the basalt parent material found in both
drainages created conversation about treatment results showing up in Mays if Jensen
watershed was treated. The geologic evidence and discussion resulted in Mays watershed
being selected for treatment.
Figure 11 shows the areas cut in Mays. The area uncut in 2005 represented the
land that is privately owned in this watershed. Treatment of this portion of Mays
occurred in the late 1970’s, prior to the initiation of this study. Jensen remains untreated,
serving as the long term control for the study. The purpose of the control is to provide
the statistical stability in which treatment effects can be differentiated.
44
Figure 11. Types of treatment in Mays watershed in 2005
The cut and remove area represented approximately 11 acres of bottom land.
Following the cutting of juniper in this area, it was determined that the volume of downed
juniper presented an elevated fire risk. This elevated fire risk was determined to be
extreme both to the potential loss of existing understory vegetation and scorching of the
soil surface, and the potential loss of monitoring equipment placed on the site. A strokedelimber was employed to delimb the juniper where it fell and deck the boles for removal
from the site by a rubber-tired skidder with grapple. The bole wood was sorted off site
into house logs and firewood. Material was sold to a local log home builder and
firewood retailer, respectively.
45
RESULTS
WEATHER
Automated weather stations were installed in Mays and Jensen watersheds in
November, 2004. Data collected included precipitation, temperature, relative humidity,
wind speed and direction, solar radiation and leaf wetness.
Precipitation
Figure 12 shows the precipitation totals for Mays watershed for the 3 water years
beginning October 1, 2003 and ending September 30, 2007. The figure also shows the
precipitation patterns for Barnes Station, a USGS long term weather station for the same
3 water years plus the 30 year average. Mays and Jensen consistently received more
moisture than the Barnes Station. Figures 13 shows the precipitation recorded in Jensen
for the last 3 water years.
C am p C reek W atershed
M ays P recipitation
20
18
Inches of Precipitation
16
14
12
2004 - 2005
2005 - 2006
2006 - 2007
30 yr ave. B arnes
B arn es 2004-05
B arn es 2005-06
B arn es 2006-07
10
8
6
4
2
0
-100
0
100
200
300
D ay of Year
Figure 12. Camp Creek Watershed Mays precipitation compared to Barnes Station
46
Figure 13. Camp Creek Watershed Jensen precipitation compared to Barnes Station
Table 8 shows the relationship of Mays and Barnes station for the three years.
Water year precipitation (October – September), ranged from 59 percent of average for
the Barnes Station in 2005-2006 (80 percent of Barnes 30-year average for Mays in the
same year) to 125 percent and 129 percent for Barnes and Mays respectively the previous
year (2004-2005). Since 1962, the lowest water year for Barnes was 1539 mm (6.06
inches) and the wettest year was 1983 with 6240 mm (24.57 inches).
Table 8. Annual water year precipitation for Mays and Barnes Station
Year
Station
Amount (mm)
2004-05
Barnes
Mays
Barnes
Mays
Barnes
Mays
3139
3667
4359
4491
2052
2781
2005-06
2006-07
% of 30 yr. Barnes
Station average
90
105
125
129
59
80
47
Air Temperature
Maximum temperatures ranged from 35 degrees C (95.48 degrees F) in Jensen
(2005) to 40 degrees C (104.4 degrees F) in Mays (2007). Minimums ranged from -21
degrees C (-5.68 degrees F) in Jensen (2007) to -18 degrees C (-0.16 degrees F) in Mays
(2005). Table 9 shows the minimum and maximum temperatures for the two watersheds
(2005-2007).
Table 9. Minimum and maximum temperatures for Mays and Jensen Watersheds 20052007.
Year
Watershed
Minimum Temp(°C)
Maximum(°C)
2005
Jensen
-19
35
Mays
-18
36
2006
Jensen
Mays
-19
-21
36
36
2007
Jensen
Mays
-21
-20
39
40
Wind
Wind speed was recorded up to 24 times per day (hourly). Recorded wind speed
represented the speed of the wind at the time of data acquisition and did not represent
wind speeds at any other time during a 24 hour period. Figures 14 and 15 show the
maximum wind speeds recorded for each day. Numbers may not actually represent the
“real” maximum if the actual maximum wind speed was not occurring at the time the
data were recorded.
48
Figure 14. Jensen Weather Station, maximum wind speed (mph)
Figure 15. Mays Weather Station. Maximum wind speed (mph)
Relative Humidity
Percent relative humidity is a measure of the amount of water in the air
compared with the amount of water the air can hold at the temperature it happens to be
when recorded. Figures 16 and 17 illustrate the daily variation in minimum percent
49
relative humidity. The lowest readings occurred during the summer months when
temperatures and evaporation rates were the highest.
Figure 16. Jensen Weather Station, daily minimum Relative Humidity.
Figure 17. Mays Weather Station, daily minimum Relative Humidity.
50
VEGETATION
Vegetative cover data were collected from eight 30 meter transects within each
watershed for the years 1995, 2003, 2005 and 2007. The first 3 years represented pretreatment variation over time and 2007 data represented vegetative changes which
occurred during the two growing seasons (2006 and 2007) post treatment of Mays. Data
from two transects per aspect (north, south, east and west) were averaged and compared
for treatment, year, aspect, and year by aspect effect. Basal cover was measured for the
perennial and annual grasses. Canopy cover was recorded for perennial forbs, live and
dead shrubs and trees. Analysis if variance (ANOVA) and single tailed t-tests were
conducted.
Table 10. Comparison of the differences in means between Mays and Jensen
Watersheds for Vegetative Cover before and after treatment.
Variable
Perennial grass
Perennial grass
Tree
Tree
Dead shrub
Dead shrub
Live shrub
Live shrub
Litter
Litter
Perennial forbs
Perennial forbs
Annual forbs
Annual forbs
Bare soil
Bare soil
Annual grass
Annual grass
Rock
Rock
treatment
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
*significant at alpha = 0.1
** significant at alpha = 0.05
Means
-3.6563
-3.0283
-12.6288
3.6892
-2.1713
-2.1121
0.3788
-0.5360
0.6650
2.1200
2.2675
0.6037
0.3163
0.0371
-4.8963
3.3379
1.1963
-0.4596
0.1850
0.1800
Difference
-0.628
P-value
one-tailed
0.4199
-0.532
0.0004**
-0.059
0.4658
0.915
0.2851
-1.455
0.4134
1.662
0.0126**
0.279
0.0611*
-8.234
0.1456
1.656
0.0317**
0.005
0.4934
51
Table 10 shows the significance of treatment in the vegetative parameters. The
removal of tree cover in Mays showed in the highly significant treatment effect on tree
cover. Other vegetative responses that are shown include a treatment increase in
perennial and annual forbs. No other parameter showed a treatment effect.
It was assumed that north and east aspects would be more favorable to understory
growth than south and westerly facing slopes. To test this, north and east slope data were
combined and averaged and south and westerly slope data were combined and averaged.
Table 11 shows the interaction of aspect and treatment on the various vegetative
parameters.
SOIL MOISTURE AND SOIL TEMPERATURE
Soil moisture probes were installed at 2 sites (upper and lower) of each watershed.
Probes were located at soil depths (profiles) of .2m (top), .45m (middle) and .76m
(bottom). This provided for a total of 6 probes per profile depth in each watershed (a
total of 24 soil moisture probes per watershed). Probe readings were recorded 4 times per
day and data were transmitted via satellite radio phone and logged on computer hard
drives. Probes were laboratory calibrated but were not field calibrated. Numbers may
not reflect actual soil moisture but provided a relative value for which trends in soil
drying and wetting could be observed.
Analysis of data involved looking at the differences between each of the
watersheds within a year and the differences between each of the treatment years.
Combinations of the pretreatment (2005-06) and post-treatment years (2006-07) for each
profile location and watershed were also analyzed.
52
Table 11. Comparison of before and after treatment means of the vegetative cover
values for Mays and Jensen watersheds for the NE and SW aspects.
Aspect
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
SW
Variable
Perennial grass
Perennial grass
Tree
Tree
Dead shrub
Dead shrub
Live shrub
Live shrub
Litter
Litter
Perennial forbs
Perennial forbs
Annual forbs
Annual forbs
Bare soil
Bare soil
Annual grass
Annual grass
Rock
Rock
Perennial grass
Perennial grass
Tree
Tree
Dead shrub
Dead shrub
Live shrub
Live shrub
Litter
Litter
Perennial forbs
Perennial forbs
Annual forbs
Annual forbs
Bare soil
Bare soil
Annual grass
Annual grass
Rock
Rock
treatment
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
After
Before
*significant at alpha = 0.1
** significant at alpha = 0.05
Mean
0.9900
0.4086
-7.2175
-4.3606
-2.4850
-1.6753
3.4500
-0.2083
-7.4350
-8.4622
2.3525
0.5469
0.1525
-0.0117
-2.2150
3.9342
0.0000
0.0000
-0.3575
-0.1158
-8.3025
-6.4653
-18.0400
11.7389
-1.8575
-2.5489
-2.6925
-0.8636
8.7650
12.7022
2.1825
0.6606
0.4800
0.0858
-7.5775
2.7417
2.3925
-0.9192
0.7275
0.4758
Difference
0.581
P-value onetailed
0.3360
-2.857
0.2619
-0.810
0.2686
3.658
0.1039*
1.027
0.4679
1.806
0.0475**
0.164
0.3510
-6.149
0.3466
0.0
-0.242
0.3122
-1.837
0.3519
-29.77
0.0109**
0.685
0.2739
-1.829
0.2255
-3.937
0.3308
1.522
0.1027*
0.394
0.0091**
-10.319
0.2001
3.312
0.0889*
0.252
0.3328
53
Average daily moisture was then calculated by adding all the readings for each
day together and dividing by the number of readings. Figures 18 through 23 show the
lowest reading of each probe for the 3 years. Probes 1 through 3 were located at the
monitoring site at the lower end of each of the watersheds and probes 4 through 6 were
located at the upper site of each watershed. Probe locations can be found in Appendix F.
Jensen Watershed
Soil Moisture - Lowest reading of the year
Bottom Profile
50
45
2005-2006
2006-2007
2007-2008
Percent Moisture
40
35
30
25
20
15
10
5
0
0
1
2
3
4
5
6
Probe
Lower Site
Figure 18.
Upper Site
Jensen Watershed Soil Moisture, end of season
probe readings for bottom profile.
Mays Watershed
Soil Moisture - Lowest reading of the year
Bottom Profile
50
45
2005-2006
2006-2007
2007-2008
Percent Moisture
40
35
30
25
20
15
10
5
0
0
1
2
3
4
5
6
Probe
Lower Site
Figure 19.
Upper Site
Mays Watershed Soil Moisture, end of season
probe readings for bottom profile.
54
Jensen Watershed
Soil Moisture - Lowest reading of the year
Middle Profile
50
45
2005-2006
2006-2007
2007-2008
Percent Moisture
40
35
30
25
20
15
10
5
0
0
1
2
3
4
5
6
Probe
Lower Site
Figure 20.
Upper Site
Jensen Watershed Soil Moisture, end of season
probe readings for middle profile.
Mays Watershed
Soil Moisture - Lowest recorded moisture
Middle Profile
50
45
2005-2006
2006-2007
2007-2008
Percent Moisture
40
35
30
25
20
15
10
5
0
0
1
2
3
4
5
6
Probe
Lower Site
Figure 21.
Upper Site
Mays Watershed Soil Moisture, end of season
probe readings for middle profile.
55
Jensen Watershed
Soil Moisture - Lowest reading of the year
Top Profile
50
45
2005-2006
2006-2007
2007-2008
Percent Moisture
40
35
30
25
20
15
10
5
0
0
1
2
3
4
5
6
Probe
Lower Site
Figure 22.
Upper Site
Jensen Watershed Soil Moisture, end of season
probe readings for top profile.
Mays Watershed
Soil Moisture - Lowest reading of the year
Top Profile
50
2005-2006
2006-2007
2007-2008
45
Percent Moisture
40
35
30
25
20
15
10
5
0
0
1
2
Lower Site
Figure 23.
3
Probe
4
5
6
Upper Site
Mays Watershed Soil Moisture, end of season
probe readings for top profile.
Figures 24 through 26 show the average soil moisture readings for each of the probe
locations by watershed. A review of these figures showed that there were recorded
differences in the averages of soil moisture at the end of the season. The end of season
soil moisture was defined as the point at which the decline of soil moisture stopped and
56
soil moisture storage began. This period usually occurred in fall or early winter with
some probes measuring their lowest reading as late as February of the water year. The
bottom profile (Figure 24) shows that the recorded soil moisture for Mays was slightly
higher in 2005 when compared to Jensen but the differences in 2006 and 2007 became
much larger. For the top and middle profiles (Figures 25 and 26), Jensen recorded a
slightly higher reading (but not significant) than Mays.
30
Percent Moisture
25
Average end of year soil moisture
readings for Jensen and Mays Watersheds
Bottom Probe
Jensen
Mays
20
15
10
5
0
2005 - 2006
2006 - 2007
2007 - 2008
Water Year
Figure 24.
Average end of season soil moisture reading for
Jensen and Mays watersheds, bottom profile.
57
Average end of year soil moisture for
Jensen and Mays Watersheds
Middle Profile
30
Jensen
Mays
Percent Moisture
25
20
15
10
5
0
2005 - 2006
2006 - 2007
2007 - 2008
Water Year
Figure 25.
Average end of season soil moisture reading for
Jensen and Mays watersheds, middle profile.
Average end of year soil mositure
for Jensen and Mays Watersheds
Top Profile
25
Jensen
Mays
Percent Moisture
20
15
10
5
0
2005 - 2006
2006 - 2007
2007 - 2008
Water Year
Figure 26.
Average end of season soil moisture reading for
Jensen and Mays watersheds, top profile.
58
T-test analysis indicated that there was no significant difference in soil moisture at
any soil depth in 2005 (the pretreatment year) between the two watersheds. Table 12
provides the average soil moisture readings for each of the 3 years.
Table 12.
Bottom
Middle
Top
Mean soil moisture readings for the end of the year by probe location
for both Mays and Jensen watersheds.
2005-06
Jensen
2006-07
2007-08
15.31
25.24
20.97
16.49
25.04
21.49
17.04
25.68
22.99
2005-06
Mays
2006-07
2007-08
20.56
21.59
17.29
26.36
23.74
19.35
25.07
24.26
19.45
Throughout the analysis of soil moisture data, the Satterthwait method (SAS protocols)
was used for combining the variances from data of the two watersheds. Significant
difference at the Alpha=.1 level did exist for the bottom profile in both 2006 and 2007
(the post-treatment years) between the two watersheds (Table 13).
Table 13.
Year
A comparison of the probe depth mean differences between
Mays and Jensen for each year.
2005
Bottom
Middle
Top
Diff. of
Mean
5.29
- 3.65
- 3.68
2006
Bottom
Middle
Top
9.87
- 1.3
- 2.14
.0540*
.1957
.4696
8.03
- 1.42
- 3.54
.1066*
.7892
.3144
2007
Profile location
Bottom
Middle
Top
* = significant at the alpha = .1
P-value
.1349
.5071
.3378
59
A t-test analysis comparing each of the treatment years against 2005 showed that
the relationship between watersheds and the treatment was strongest in 2006
(precipitation was 125 – 129 percent of normal) (Table 14). A mean increase in the
difference was greatest for the bottom profile (significant at alpha = .1). All other probe
locations showed a slight increase in the difference indicating that end of season soil
moisture had increased more in Mays than in Jensen for each of the post treatment years
but that difference was not significant.
Table 14. A comparison of the mean differences between each of the treatment
years and the pre-treatment year.
Year
Profile Location
2006 vs. 2005
2007 vs. 2005
Bottom
Bottom
Diff. of
Means
4.58
2.74
P-value
2006 vs. 2005
2007 vs. 2005
Middle
Middle
2.35
2.23
.1957
.2283
2006 vs. 2005
Top
2007 vs. 2005
Top
* significant at Alpha = .1
1.54
0.42
.3421
.8085
.0783*
.1637
T-test analysis comparing the average of 2006 and 2007 against 2005 again shows
That the average for treatment years was significant at the alpha = .1 level for the bottom
profile. The other profiles showed slight increases in overall moisture but were not
significant (Table 15).
60
Table 15.
A comparison of the average of the mean differences for the two
post-treatment years with the pre-treatment year.
Year
Profile Location
2006-07 vs. 2005
2006-07 vs. 2005
2006-07 vs. 2005
Bottom
Middle
Top
Jensen
1.45
0.12
1.27
Diff. of
Means
Mays
5.15
2.44
2.14
P-value
Diff.
3.7
2.29
0.87
.1002*
.1796
.6132
significant at Alpha = .1
The time of year in which the lowest reading was acquired was examined. The
purpose of this was to see if there were differences in when the lowest reading occurred
and if there was any relationship to the treatment. Table 16 illustrates day of year when
the lowest reading occurred. Day of year was calculated from September 1 of each year.
September 1 was used so that no negative numbers existed. Table 17 shows the average
number of days for each profile location by year for each of the watersheds. Changes in
difference in number of days when soil moisture began to accumulate could be affected
by vegetation (tree intercept) and/or the site’s ability to encourage infiltration.
T-test showed a significant difference within each year between the watersheds
for some profiles within each of the years (Table 18). When the difference in means was
negative, that showed that Mays took fewer days (from September 1) to begin to store
soil moisture. In the treatment year (fall 2005), the bottom profile was significant and
corresponded to cutting of the trees. In 2006 and 2007, the top (alpha = .01) and middle
61
(alpha = .05) profiles were significant. In 2007, the bottom profile also was again
significant (alpha = .01) indicating that with tree removal the site’s ability to begin
storing soil moisture was improved with treatment.
In a comparison of treatment years with the pre-treatment readings, the top profile
was the only profile to show significant difference. T-test showed significant difference
(alpha = .01) for the top profile when comparing the first treatment year 2006 and 2005
but no significant difference for the second treatment year 2007 and 2005 (Table 19).
This reflected the dry year of 2007 (80 percent of average) with 2005 which was near the
long term precipitation average for Barnes Station.
A t-test looking at the average of the two post treatment years compared to the
pretreatment year is found in Table 20. Significance was only found in the changes in
the top profile (alpha = .01). A reduction in the number of days for the middle profile
was also found but was not significant. There was no difference in the bottom profile
when comparing Mays to Jensen.
62
Table 16. Number of days from Sept 1 in which lowest soil moisture reading was
recorded.
2005-06
Watershed
Mays
Probe
Top
Site
Lower
Upper
Middle
Lower
Upper
Bottom
Lower
Upper
Jensen
Top
Lower
Upper
Middle
Lower
Upper
Bottom
Lower
Upper
2006-07
2007-08
Date
11/7/2005
11/8/2005
9/24/2005
11/17/2005
12/19/2005
11/10/2005
Days
68
69
24
78
110
71
Date
10/31/2006
11/1/2006
11/1/2006
12/15/2006
1/15/2007
12/4/2006
Days
61
62
62
106
137
95
Date
10/6/2007
10/22/2007
10/6/2007
11/15/2007
9/29/2007
*************
Days
36
52
36
76
29
11/29/2005
12/13/2005
11/5/2005
11/29/2005
11/6/2005
12/13/2005
90
104
66
90
67
104
11/19/2006
2/3/2007
11/18/2006
11/2/2006
11/18/2006
1/18/2007
80
156
79
63
79
140
11/15/2007
12/18/2007
11/8/2007
12/18/2007
11/17/2007
12/17/2007
76
109
69
109
78
108
11/26/2005
12/19/2005
12/11/2005
12/3/2005
11/4/2005
11/21/2005
87
110
102
94
65
82
2/19/2007
1/9/2007
11/10/2006
12/16/2006
11/1/2006
2/2/2007
172
131
71
107
62
155
12/9/2007
12/10/2007
12/4/2007
12/20/2007
12/14/2007
12/18/2007
100
101
95
111
105
109
11/10/2005
11/6/2005
11/9/2005
11/3/2005
11/10/2005
11/4/2005
71
67
70
64
71
65
2/2/2007
2/2/2007
2/2/2007
2/2/2007
2/3/2007
2/2/2007
155
155
155
155
156
155
11/15/2007
10/18/2007
11/4/2007
11/15/2007
11/15/2007
11/4/2007
76
48
65
76
76
65
12/14/2005
12/22/2005
1/22/2006
11/11/2005
11/11/2005
************
105
113
144
72
72
1/30/2007
2/2/2007
2/3/2007
12/9/2006
2/3/2007
2/2/2007
152
155
156
100
156
155
12/24/2007
12/25/2007
12/28/2007
11/30/2007
12/24/2007
12/23/2007
115
116
119
91
115
114
12/20/2005
12/19/2005
12/3/2005
12/30/2005
12/16/2005
12/17/2005
111
110
94
121
107
108
2/1/2007
2/1/2007
2/2/2007
1/23/2007
12/24/2006
2/3/2007
154
154
155
145
115
156
12/26/2007
12/19/2007
12/21/2007
12/23/2007
12/18/2007
12/21/2007
116
110
112
114
109
112
******* = no probe reading during this time period
63
Table 17.
The average number of days (from September 1) until soil moisture
began to accumulate at the various probe depths for each of the
watersheds.
Watershed
Profile
2005-06
2006-07
2007-08
Mays
Top
Middle
Bottom
70
87
90
87
100
116
46
92
104
Jensen
Top
Middle
Bottom
68
101
109
155
146
147
78
112
112
Table 18. A comparison of the average number of days between Mays and Jensen
watersheds when soil moisture began to accumulate for each year.
Year
Profile Location
Mays
Watershed
Jensen Diff.
P-value
2005-06
Bottom
Middle
Top
90
87
70
109
101
68
- 19
- 14
2
.0311**
.3462
.8631
2006-07
Bottom
Middle
Top
116
100
87
147
146
155
- 31
- 46
- 68
.1491
.0293**
.0003***
2007-08
Bottom
Middle
Top
104
92
46
112
112
68
-8
- 20
- 22
.0086***
.0456**
.0398**
** significant at Alpha = .05
*** significant at Alpha = .01
64
Table 19. A comparison of the difference in the average number of days before soil
moisture accumulation began by profile depth for each of the treatment years
and the pre-treatment year between Mays and Jensen.
Year
Profile Location
2006 vs. 2005
Bottom
Middle
Top
Mays
26
13
17
2007 vs. 2005
Bottom
Middle
Top
14
5
- 24
Watershed
Jensen
38
45
87
Diff.
- 12
- 32
- 70
P-value
.5688
.1114
.0001***
3
11
10
11
-6
-34
.2735
.6433
.8085
*** significant at Alpha = .01
Table 20. A comparison of the difference of the average number of days before
soil moisture accumulation began for the combination of post-treatment years
with the pre-treatment year.
Year
Profile Location
Mays
2006-07 vs 2005
Bottom
Middle
Top
20
9
- 12
Watershed
Jensen Diff.
20
28
48
0
- 19
-60
P-value
.9403
.1796
.0006***
*** significant at Alpha = .01
GROUND WATER
Depth to water was recorded since January, 2003. In June, 2005, data collection
also included automated recordings which were collected 4 times per day. Analysis was
done on days of recorded water. Figure 27 illustrates the annual variation in the 4 years
65
of data collecting for both watersheds. Table 21 shows the average days pre and post
treatment for the two watersheds and the differences.
Days of recorded water in wells
400
350
300
250
03-04
04-05
200
05-06 Treatment Year
06-07
150
100
50
0
Well Well Well Well Well
1
2
3
4
5
Well
6
Well
1
Mays
Well Well Well Well Well
2
3
4
5
6
Jensen
Wells
Figure 27.
Days of recorded water in wells for the
water years 2003-04 through 2006-07.
Table 21. Comparison of average number of days water was recorded in each well. Pre
and post treatment years consist of 2 years each.
Watershed
Mays
Well
1
2
3
4
5
6
Pretreatment
112.5
119.5
195.5
195.5
156
269.5
Posttreatment
128.5
135
285
209
197
342.5
Difference
16
15.5
89.5
13.5
41
73
Jensen
1
2
3
4
5
6
70
78.5
283.5
314.5
283.5
167.5
82
89
296
361.5
296
141
12
10.5
12.5
47
12.5
-26.5
66
Figure 28 illustrates the average of each of the 2-year periods of pre and post
treatment. The difference between the two years is shown as well.
Days of recorded Well Water
2nd year after treatment
400
350
300
250
Pretreatment Average
200
2 yr.ave Post Treatment
150
Difference
100
50
0
-50
Mays
Jensen
Wells
Figure 28.
Days of recorded well water pre and post treatment and
the associated difference.
A two-sample t-test was applied to test for treatment differences. Results of that
test are in Table 22. A significant difference in the change of number of days water was
present in Mays following treatment was at the alpha > .05.
Table 22. Test for treatment differences in days of recorded water.
Watershed
Jensen
Mays
Average
Difference
11.33
41.42
** significant at alpha = 0.05
Standard
Error
11.63
11.63
t value
-1.83
P-value
one-tailed
0.049**
67
Figures 29 and 30 illustrate the changes in curve shape and depths that water was
recorded over time. The curve shape for Jensen (control) was similar for all years. The
height of curve (closeness of water to the surface) was a reflection of the precipitation
received during that water year. The curve shape for Mays changed between the
pretreatment years and the post treatment. The pretreatment curve for Mays was best
described as spiked, with the peak of the curve occurring quickly and then dropping off.
The post treatment curves were flatter, with the peaks not being as sharp. Since the
treatment, well 6 in Mays did not go dry. Well 4 in Jensen recorded water in years 2004,
2005 and 2006 but went dry in late September, 2007.
Depth to Ground Water - Jensen
Well 1
Well 2
Well 3
10
Well 4
Well 5
Well 6
Depth (fee
0
-10
-20
-30
03 004 004 004 004 004 004 005 005 005 005 005 005 006 006 006 006 006 006 007 007 007 007 007 007 008
20
/2
/2
/2
/2
/2
/2
/2
/2 5/2 5/2 5/2 5/2
/2
/2
/2
/2
/2
/2
/2
/2
/2 5/2 5/2 5/2
/2
5/
1
1
1
15 /15
15 /15 /15 /15 /15 /15
1
1
15
15 /15 /15 /15 /15 /15
15 /15 /15
1
/1
/1
2
8
2/
2/
4
4
6
6/
8/ 10/
2/
8
4
2/
6
4
8/ 10/ 12/
6
2/
1
10
12
12
10
12
Figure 29.
Depth to ground water, Jensen wells 2003 – 2007.
Depth to Ground Water - Mays
Well 1
Well 2
Well 3
Well 4
Depth (fee
10
Well 5
Well 6
0
-10
-20
12
/1
5/
20
0
2/
15 3
/2
0
0
4/
15 4
/2
00
6/
15 4
/2
00
8/
15 4
/2
10 00
4
/1
5/
2
12 00
4
/1
5/
20
0
2/
15 4
/2
4/ 005
15
/2
6/ 005
15
/2
00
8/
15 5
/
10 200
5
/1
5/
2
12 00
5
/1
5/
2
2/ 005
15
/2
4/ 006
15
/2
00
6/
15 6
/2
0
0
8/
15 6
/2
10 00
6
/1
5/
12 200
6
/1
5/
20
06
2/
15
/2
4/ 007
15
/2
00
6/
15 7
/2
8/ 007
15
/2
10 00
7
/1
5/
2
12 00
7
/1
5/
2
2/ 007
15
/2
00
8
-30
Figure 30.
Depth to ground water, Mays wells 2003 – 2007.
68
SPRING FLOW
Flows from two spring boxes, one in Mays and one in Jensen were measured
approximately every two weeks since September, 2004. Measurements were automated
in May, 2005 and thereafter were recorded 4 times per day. Manual measurements were
taken throughout the entire period for calibration and verification purposes. Data loggers
were not able to accurately measure flows below 3 gpm.
Figure 31 illustrates the differences in output between the two springs and the
differences between years. Spring flow was dependent on timing/type and amount of
precipitation. Base flow (that flow which is least likely to be influenced by a recent
precipitation event or snow melt period) was the late season flow, the period between
July and November.
Spring Flow
Mays
60
Jensen
40
30
20
10
Date
Figure 31. Mays and Jensen spring flow, 2004 – 2007.
12/11/2007
10/29/2007
9/10/2007
7/16/2007
5/17/2007
3/29/2007
2/19/2007
1/2/2007
10/30/2006
9/5/2006
7/5/2006
5/8/2006
2/26/2006
1/13/2006
10/5/2005
11/28/2005
8/11/2005
6/13/2005
5/2/2005
3/16/2005
2/8/2005
11/29/2004
0
9/28/2004
Gallons per Minu
50
69
Figure 32 shows the base flows for the 2 years prior to treatment and the 2 years post
treatment.
Late Season Spring Flow
and
Monthly Precipitation
9/1/03
25
1/1/04
5/1/04
9/1/04
1/1/05
5/1/05
9/1/05
1/1/06
5/1/06
9/1/06
1/1/07
5/1/07
9/1/07
6
Mays (treated)
Jensen (control)
Monthly Precipitation
4
Gallons per Minute
2
0
15
-2
10
-4
-6
Inches of Precipitation
20
5
-8
0
9/1/03
-10
1/1/04
5/1/04
9/1/04
1/1/05
5/1/05
9/1/05
1/1/06
5/1/06
9/1/06
1/1/07
5/1/07
9/1/07
Date
Figure 32.
Late season flows for Mays and Jensen springs.
Table 23 shows the t-test results for comparisons of late season flow (lowest flow
recorded) between the two watersheds and the pre and post treatment years. The one
tailed P-value is significant at alpha = .05.
Table 23.
T-test for spring flow data, lowest flow recorded (gpm)
Treatment
Year
Pre
Pre
Post
Post
2004
2005
2006
2007
**significant at Alpha = 0.05
Watershed
Difference
Mays
Jensen
1.87
0.20
1.67
1.90
0.13
1.77
4.80
0.23
4.57
3.6
0.00
3.60
Difference
Standard error
t-test
One tailed P-value
Mean Variance
1.720 0.00500
4.085 0.47045
2.365
0.4876
4.8506
0.020 **
70
CHANNEL FLOW
The channel flows of Mays and Jensen were ephemeral in nature, only existing
during periods of snow melt or summer thunderstorm activity. In an attempt to look at
channel flow as influenced by the treatment, total flow volumes and days of flow were
looked at for the entire year and seasonally looking at spring flows (Figure 33). Spring
flows were singled out because they were most likely to occur. For the years of data
collection in which complete years were recorded (10 out of 13 years) Mays had 6 years
of springtime flow data and Jensen had 7 years. Spring flow was defined as any flow
occurring from January 1 to June 30 of each year. In all but 1 year (1996) days of
channel flow in Mays exceeded Jensen.
Days of Springtime Channel Flow
180
Mays
Jensen
160
140
Days of Flow
120
100
80
60
40
20
0
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
Year
Figure 33.
Number of days of channel flow from January to June.
Figure 34 shows the springtime flow in cfs (cubic feet per second). Unlike the
days of flow, Mays exceeded flow in Jensen in all years except 3. One of those years it
did not exceed Jensen was in 2007, the 2nd year of post treatment.
71
Total Springtime flow in Cubic feet per second
35
Mays
Jensen
30
25
CFS
20
15
10
5
0
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
Year
Figure 34.
Springtime channel flow by year (January
through June) in cfs.
Figure 35 shows days of annual channel flow. This figure represents the total
number of days that flow was recorded. As with the springtime channel flow, Mays flow
days were always greater than Jensen except for one.
Days of Annual Channel Flow
180
Mays
Jensen
160
140
Days of Flow
120
100
80
60
40
20
0
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
Year
Figure 35.
Number of days of annual channel flow.
Figure 36 illustrates the highly variable annual channel flow associated with the
two watersheds. In 1997, total flow of the 2 watersheds was almost identical (5.99 cfs for
Mays and 5.65 cfs for Mays). In 2004, springtime flow for Jensen was greater than Mays
72
(12.95 cfs vs. 10.54 cfs), but total annual flow flipped with Mays ending up with greater
annual flow (13.15 cfs vs. 12.95). While springtime flow, 3 years of Jensen flow
exceeded Mays, in annual flow, only 2007 did Jensen’s total flow exceed Mays.
Total flow in cubic feet per second
35
Mays
Jensen
30
25
CFS
20
15
10
5
0
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
Year
Figure 36.
Total annual flow in cubic feet per second
2008
73
DISCUSSION
WEATHER
The water balance equation is made of inputs (rain and snow) and outputs.
Outputs are interception and evaporation, soil infiltration (for plant growth and ground
water recharge) and overland flow. Weather in its many forms (precipitation,
temperature, wind and relative humidity) plays an important role in the function of semiarid watersheds. Weather interacts with vegetation and the soil surface. Precipitation
interacts by way of storm intensity, total amount per event, frequency of event, and type
of precipitation. Temperature affects the type of precipitation, the condition of the soil
surface (frozen, dry) and influences plant utilization via transpiration. Wind affects
potential soil surface and plant water utilization by increasing evaporation as wind speed
increases. Wind may also affect deposition of precipitation, particularly in the winter
when blowing snow may be swept from one site and deposited on another. Relative
humidity is important in controlling the rate of transpiration by creating a moisture deficit
in the atmosphere surrounding the plant. As the deficit gets larger, transpiration increases
until the plant can no longer move adequate amounts of water at which point the plant
will close its stomata if possible. If the plant cannot control its water use this way, the
plant will desiccate.
Precipitation
The annual precipitation for the watersheds is 3500 mm or 13.75 inches (Barnes
Station, 30 year average). Within the watersheds, precipitation for the last 3 water years
has varied from 80 to 129 percent of the average. Timing of precipitation is critical to
74
plant growth and ground water recharge. Sixty to Sixty-five percent of the total is
delivered from October 1 to May 1 in the form of snow or rain.
Typically the site will receive fall rains, wetting the soil surface. Precipitation
amounts from October 1 to December 1 ranged from 381 to 1016 mm during the study
period. During the winter months, precipitation generally comes in the form of snow.
Snow depths were variable during the study time. During the study, there were winters
of relatively no snow pack (bare ground observed most of the winter) to winters with
snow pack depths of 3048 mm (12 inches) to 4064 mm (16 inches). During the winters
of 2005-2006 and 2007–2008, soil temperatures in both watersheds stayed above 0
degrees Celsius (32 degrees Fahrenheit). Snow pack during these two winters lasted into
mid-March, with peak depths of 4000 mm plus. During the winter of 2006–2007, snow
depth was relatively nonexistent and February soil temperatures approached -5 degrees
Celsius (22 degrees Fahrenheit). It was during this period in February, 2007 that the
watersheds experienced a rain on frozen ground event that resulted in significant overland
and channel flow in both watersheds.
Summer thunderstorms can contribute significant amounts of precipitation to the
total annual precipitation amount but are more localized and sporadic in nature. It is not
uncommon to see that a storm came through and impacted part of a watershed but missed
another part of the same watershed. Sometimes a storm would completely miss one of
the watersheds. On June 29, 2006, a summer thunderstorm occurred on both watersheds.
Precipitation amounts recorded equaled 177 and 203 mm (152 and 177 mm in the first
hour) respectively in Mays and Jensen. Flow (10.2 cm on the staff gauge) was recorded
in Jensen’s flume, the untreated control, but not in Mays, the treated watershed. An
75
inspection of each of the watershed’s channels revealed that large amounts of overland
flow had deposited sediment in Jensen (burying three-fourths of cross-section number 4
and filling the flume) while the channel in Mays showed no signs of overland flow and
sediment movement (even though this was only 6 days after finishing the skidding
operation in Mays, which removed the bole wood from 4.45 hectares [11 acres]
surrounding the channel).
As a result of these observations and measurements, it was apparent that
precipitation as well as organic ground cover played a vital role in how these watersheds
functioned. Storm intensity and type of precipitation are some of the critical factors in
determining how a watershed will respond to precipitation inputs. Snow cover is critical
in controlling winter time soil temperature, the degree in which the soil is frozen and as a
result, whether that soil will accept infiltration or result in overland flow.
Air Temperature
Air temperature affected the type of precipitation delivered to the study area.
Cold temperatures will usually result in precipitation coming in the form of snow and
with warmer temperatures, coming in the form of rain. Without snow cover, cold winter
temperatures drive soil temperatures well below freezing. Frozen soils act like a lid on
the soil surface and prevent water infiltration.
Air temperature also affects plant growth. As winter turns to spring and then to
summer, warmer air temperatures promote photosynthesis and plant growth. Plant
growth utilizes soil moisture. Vegetation serves as a thermal blanket, dampening the
heating of the soil surface by intercepting short wave radiation.
76
Table 9 (pg. 47) illustrates the maximum and minimum temperatures recorded in
the two watersheds. There was less than 1 degree Celsius (2 -3 degrees F) difference in
either the winter lows or the summer highs.
Wind
Wind can increase a plant’s water use by drying the surface of the leaf tissue,
creating a moisture imbalance, which results in an increase in water demand by the plant.
No measurable differences in recorded wind speed were observed as a result of the
treatment. In each watershed, the wind speed indicator was located on top of a 4.5 m
pole. At this height, the indicator was generally above the associated plant community
and most importantly, the height of the tree canopy. Following tree removal, one might
suppose that wind speeds closer to the soil surface would increase. However, this
component was not measured.
There was some concern that the removal of the trees would affect snow
deposition as a result of drifting. In looking for evidence of drifting snow within the
juniper stand or in the treated area, little evidence of drifted snow was observed. During
the winter of 2007-08, some small drifts were observed in the cut areas of Mays. Drifts
were found on the northwest slopes (winds predominately out of the southwest) behind
sagebrush. However, the drifts were small and were most easily observed only as the sun
was setting and shadows were cast. Drifted snow did not appear to be a significant factor
in winter snow deposition in these watersheds.
77
Relative Humidity
Relative humidity is a relationship between the amount of moisture held in the air
and the amount of moisture the air can hold at a given temperature. It is a relationship
that is largely connected to air temperature and thus is a landscape controlled feature, one
not easily affected by local (watershed) conditions. A review of the data showed no
indication that relative humidity was affected by the presence or absence of western
juniper in the watersheds.
VEGETATION
Initial changes in vegetation showed some statistical significance as a result of the
treatment. Significant changes attributed to the treatment included a decrease in tree
cover, and an increase in perennial and annual forbs cover as well as annual grass cover.
Vegetative changes that seem important but were not statistically significant were a
decrease in percent bare soil and a slight increase in live shrub cover. Perennial grass
cover showed a slight decrease. These changes do not seem surprising. The understory
was present in the treated watershed before the removal of the trees. Perennial grass
cover was measured as basal cover so these plants were already on site. There was no
evidence or expectation of new grass plants during the first two years following the
cutting of the junipers. Understory plants may have been suppressed and the increase in
perennial forb and shrub cover would support this concept. Perennial grass cover
measured an insignificant 0.05 percent reduction in cover from pretreatment levels. This
is well within the limits of observer error.
78
What was evident but not shown in the data was the large amount of reproduction
culms associated with the perennial grasses. Observer notes indicated that in Mays
watershed, for bluebunch wheatgrass, Junegrass, bottlebrush squirrel tail there were 10’s
of reproductive culms per plant. Idaho fescue had generally less than 10 per plant. In
Jensen, reproductive culms were limited, with the major perennial grass species all
having less than 10 culms per plant.
Greenness of the plants was also observed during 2007 with the majority of grass
plants (greater than 70 percent) in Mays still having green leaves in the plant bunch at the
time of observation (July 25, 2007) while those same species of grass plants in Jensen
were generally dormant and not showing any green (less than 20 percent of the plants).
This observation was made in a year when annual precipitation was less than 80 percent
of the long term average for the site.
Changes in plant community were also examined based on aspect (Table 11, pg.
52). It was thought that generally, north and east slopes would be less hostile to plant
growth while south and west slopes are more hostile due to orientation to the sun and the
associated warming of the site and available soil moisture. A review of the analysis
showed that on north and east slopes, only perennial forb cover increase was significant.
Non-significant changes associated with aspect included an increase in perennial grass
cover, increase in live shrub cover and an increase in bare ground.
Significant changes on combined south and west slopes included an increase in
annual grass and perennial/annual forb cover. Other vegetative parameters showed only
non-significant changes.
79
Vegetative changes are anticipated to be much greater in the years to come.
Following the large amount of seed produced in 2007, it is anticipated that within Mays
watershed, bare ground will continue to decrease and forb and grass cover will increase
as new plants are produced and existing plants continue to be released. Shrub cover will
increase in the coming years primarily from release but new plants may increase as seed
production increases as well.
SOIL MOISTURE AND SOIL TEMPERATURE
A review of the soil moisture data provided an opportunity to see how
precipitation moved into the system and was utilized. The data provided an illustration
of relative soil moisture over time and showed how the soil wets and dries, season to
season, year to year. What was most interesting was the evidence (Figure 24, pg. 56) that
by removing the deep rooted juniper, deep soil moisture was no longer being utilized for
the entire length of the growing season and therefore the treated watershed ended the
season with more soil moisture than where it had the year before. While it is too early to
know if this is a long term trend, it is one that showed itself not only in the above average
precipitation year of 2006 (129 percent of average) but also in the dry year of 2007 (80
percent of average).
In a functioning watershed, the first thing that happens is the capture of the water
droplet where it falls. Vegetation (plants and litter) serves to break up the speed of the
droplet before it strikes the soil. As quickly as possible, that water droplet should
infiltrate into the soil profile. Much like a sponge, soils that have some moisture in them
are more likely to absorb additional moisture quicker than dry soils. Dry soils can be
80
hydrophobic and almost always repel water at first. Soil organic matter increases a soil’s
ability to store water. Growing plants increase soil organic matter through root
production followed by root die off and decomposition. Increased soil moisture can
increase plant root productivity if it is available during the growing season. The removal
of the juniper reduced water utilization during periods when the other components of the
plant community were not growing. The removal of juniper allowed soil moisture to
build during the winter and early spring period providing the opportunity for improved
herbaceous plant productivity and potential ground water recharge.
A review of Figures 18 through 23 (pgs. 53-55) showed how end of season soil
moisture amounts compare at the various probe depths. For Jensen (control), the average
readings for both sites showed that for the 3 years, end of season soil moisture was
relatively constant for both the bottom probes and those at the middle profile. The top
profile probes showed a slight but insignificant increase in end of season soil moisture
level in 2007. For Mays, all soil profiles (bottom, middle and top) showed increases in
end of season soil moisture levels as a result of the treatment. The bottom probes,
averaged across both sites in Mays watershed, were the only probe location where the
observed change was significant as a result of the treatment (alpha = .1).
Soil temperatures were influenced by conditions near the soil surface (Figures 37
and 38). In the winter, the presence or absence of a snow pack controls whether or not
subsurface soil temperatures drop below 0 degrees Celsius (32 degrees Fahrenheit).
During the winters of 2005-06 and 2007-08, near surface soil temperatures did not drop
below 0 degrees Celsius.
81
Mays Upper Soil Site
Average Daily Temperature
Degrees Fahrenheit
90
80
70
60
50
40
30
20
10/1/07
8/1/07
6/1/07
4/1/07
2/1/07
12/1/06
10/1/06
8/1/06
6/1/06
4/1/06
2/1/06
12/1/05
10/1/05
8/1/05
0
6/1/05
10
Date
Figure 37. Average daily soil temperature, Mays upper monitoring site.
While ambient air temperatures dropped to near -17 degrees Celsius (0 degrees
Fahrenheit), the snow pack served as an insulator, moderating soil temperatures. In
contrast, during the winter of 2006-07, soil temperatures approached -5 degrees Celsius
(22 degrees Fahrenheit) for an extended period of days. Frozen soil prevents snow melt
Jensen Upper Soil Site
Average Daily Temperature
Degrees Fahrenheit
90
80
70
60
50
40
30
20
10
Date
Figure 38. Average daily soil temperature. Jensen upper monitoring site.
10/1/07
8/1/07
6/1/07
4/1/07
2/1/07
12/1/06
10/1/06
8/1/06
6/1/06
4/1/06
2/1/06
12/1/05
10/1/05
8/1/05
6/1/05
0
82
and early spring rains from entering the soil profile. On frozen soil, as rains occur and
snow melts, overland flow is observed. During the winters with adequate snow pack, and
therefore unfrozen soils, overland flow was not observed and channel flow was
originating from bank seepage.
During the summer months, there did appear to be a slight elevation in soil
temperatures in Mays watershed following treatment. Soil temperatures during the
summer months for both watersheds were similar the year prior to treatment (2005).
Following the cutting of the trees, there appeared to be slight increase in summer time
soil temperatures in Mays in both 2006 and 2007 (.5 – 1.5 degrees Celsius, 2 – 3 degrees
F). The removal of the tree canopy and the corresponding shade allowed the direct
sunlight greater access to the soil surface, causing it to warm. One might suppose that
this temperature increase would result in a shorter growing season (removal of soil
moisture through transpiration and soil water evaporation) but this did not appear to be
the case. A review of the data indicated that soil moisture was available longer in Mays
following treatment than before treatment. In reviewing other site parameters (soil
moisture, plant cover and observed greenness throughout the summer) it did not appear
that these observed temperature changes were negative.
GROUND WATER
In a healthy watershed, the storage and safe release of water is important since
water is a limited resource. From a management perspective, it becomes important to
make that water available for all beneficial uses over an extended period of time.
83
Beneficial uses include plant growth (on site), water in the stream for fisheries and other
aquatic species and irrigation/recreational uses farther down the watershed. Just like a
hydrograph for a stream system, a hydrograph for ground water shows how water moves
through the system subsurface, accumulating ground water as the snow melts and spring
rains come, reaching a peak in late spring and early summer, then diminishing throughout
the summer, disappearing by mid to late summer. For a watershed that is not functioning
properly, a hydrograph may show a spike in flow following a storm event or spring snow
melt, indicating that the site was not able to absorb what is delivered during the time it
was delivered and it immediately shows up in the channel. For a watershed that is
functioning properly, contact time between the soil and the precipitation is increased on
the uplands, increasing the opportunity for infiltration. Water that cannot infiltrate and
moves down slope to the channel as overland flow potentially becomes flood waters.
These flood waters are moved onto the floodplain, slowing it down and again increasing
the opportunity to have that water move into the soil profile through infiltration. The
hydrograph in this case would show a flow that had its peak flow dampened and its
length of flow (expressed in days) extended.
Monitoring the ground water in Mays and Jensen provided a look beneath the soil
surface and into the “channel” we cannot see. Figures 29 and 30 (pg. 67) illustrate the
hydrographs for ground water in these two watersheds. The hydrograph for Jensen
(Figure 29) exhibited a curve that has been relatively consistent for the 4 years since the
wells were placed. While height and width of the curve was somewhat controlled by the
amount and timing of precipitation for each year, the shape was a reflection of how the
“system” was working in Jensen, which included the condition of the vegetation.
84
For Mays (Figure 30) we saw the effect of the treatment. During the first two
years (prior to treatment) the hydrograph showed that most of the wells had a short,
spiked curve that reflected spring run off and a system where water moved through the
watershed very quickly. Following treatment, the next two years showed a significant
dampening of the curve (no longer spiked) and a flattened, extended flow of water
subsurface. For well 6, even more significant, was that since treatment two years ago this
well has not gone dry, whereas, before treatment it was always dry by late summer.
It is important to point out that precipitation for both years prior to treatment and
after treatment have been very similar (Figures 12 and 13, pgs. 45-46). Prior to
treatment, we had a relatively wet year (110 percent) and a relatively normal year (95
percent), and in the post treatment years, we have had a relatively wet year (125 percent)
and a relatively dry year (80 percent). The reason for pointing this out is that there is a
tendency to think that treatment effect was just because it was a “wet year”. Because the
analysis used the average of pre-treatment years and post-treatment years, the extremes
have been dampened.
Figure 27 (pg. 65) shows the number of days water was recorded in the wells for
each year. The pre-treatment normal year was 2004-05 and the post-treatment dry year
was 2006-07. A review of Figure 23 (pg. 55) indicated that five of the six wells in Mays
showed increases in the post-treatment days of recorded water and the sixth well was
equal to its pretreatment level. Within Jensen, two wells showed slight increases in days
of recorded water but four of the wells showed a clear reduction in number of days of
recorded water. A comparison of the wet years (2003-04 and 2005-06) showed that for
both Mays and Jensen, a slight to large increase was measured in all wells. Therefore,
85
water availability was significant in the dry years as a result of the treatment. During the
wet years, precipitation may overcome watershed condition and therefore a “shortage”
may not be as evident.
SPRING FLOW
A review of the spring flow data provided an insight similar to that of the well
data. Springs are a surface expression of ground water. As water moves subsurface
down slope, if it encounters an impervious obstruction, deflection from that obstruction
may force it to the surface. If this happens, a spring is formed. Anecdotal information
about juniper control includes the observation that following juniper removal, flow from
known springs increased and sometimes new springs were identified in treated areas
where they were not documented to be before.
If springs are a product of ground water, and in a functioning watershed increased
infiltration results in increased stored soil water, than increased spring flow would be an
indicator that site is functioning better. Figure 31 (pg. 68) shows the actual
measurements of each of the springs located in Mays and Jensen. The general shapes of
the curves (the peaks and flow lengths) are a reflection of the amount of precipitation
received each year. For treatment effects, what is of concern is the change in the
differences between the two.
For the purpose of minimizing the inputs of precipitation on spring output, an
analysis of late season spring flow was conducted. Late season flow was selected
because it is the flow period least likely to be influenced by a recent precipitation event
86
while illustrating any changes associated with increased infiltration of precipitation. Late
season spring flows are comparable to the “base flow” of a stream or river.
A change in the differences in the two springs output was measured. Figure 31
(pg. 68) illustrates the differences in output between the two springs both prior to and
after treatment. Analysis of this time period showed that a significant increase in the
difference in output has occurred since treatment, almost 225 percent increase in the
difference (alpha>.05). Conversely, during the dry year of 2007 (80 percent of average),
late season flow in Jensen ceased for the first time since monitoring began.
Developing a spring for livestock water or other purposes can be costly. The cost
of developing the springs in this study was $4000.00 each. Landowners and land
managers want to be sure their investment in a water development is a sound one and in
the case of a spring, that soundness is measured in flow and that flow’s ability to meet the
needs of the intended use (i.e., gallons of water per minute to meet the needs of a given
number of livestock for a given period of time). Without vegetative treatment, both of
these springs provided minimal late season flows. Given the annual precipitation of 3500
mm (13.75 inches), output from either spring (Mays or Jensen) may not have been
enough for the investment to develop. In fact, the landowner was hesitant to provide his
approval for the spring developments because the flow, as he knew it, was seasonal and
had only the potential of creating livestock behavior problems.
This study showed that the increase in output for the Mays spring following
treatment justified the investment in developing it. Even when annual precipitation was
80 percent (2800 mm or 11 inches) of the long term average, flow was greater in late
season following treatment when compared to pretreatment flows (approximately 225
87
percent – 1.8 gallons per minute to 4.2 gallons per minute). Mays spring now provides
off-site water in late August for 240 cow-calf pairs.
CHANNEL FLOW
Channel flow in the two watersheds is ephemeral. These channels only have flow
during periods of snow melt and extreme summer thunderstorm activity. Ephemeral
channels tend to be more influent in relation to the ground water, contributing to ground
water rather than ground water contributing to channel flow.
Comparisons of ephemeral channel flows or days of flow did not show a
relationship to the treatment. A comparison of Figures 34 and 36 (pgs. 71-72) illustrated
that in most years, recorded channel flow occurred during the spring and early summer
months. In 1996 and 2004 total annual days of flow (Figure 36) were greater than days
of springtime channel flow, a result of late summer thunderstorms and early fall rain
(Figure 34). In all years but one, Mays flowed longer than Jensen. Only in 1998 did
Jensen flow longer when compared to Mays. In 2007, while length of flow was greater in
Mays, Jensen’s flow as measured in accumulated cubic meters per second (cms.) was
greater than Mays’ flow.
Of special note in the observation of these systems was the winter of 2006,
following the cutting of juniper in Mays. The snow pack, which began its accumulation
in December, 2005 was static at approximately 40 cm (16 inches). December and early
January rain events saturated the snow pack. While no water content measurements were
taken, field notes indicate that the snow was saturated and frozen on top. Field notes also
indicated that the snow pack was solid enough for researchers to be able to walk on top of
the snow without breaking through. As mentioned earlier, soil temperatures during this
88
period did not drop below 0 degrees Celsius (32 degrees F) for either watershed. Channel
flow in Mays began on January 7, 2006. Flow was recorded through mid-June, 2006. In
contrast, flow in Jensen did not begin until April 1, 2006 and ceased to flow by early
May. During this period, all observations for both watersheds indicated that flow was
generated exclusively from bank seepage and that no evidence of overland flow was
observed for either watershed.
In contrast, during the winter of 2007, very little snow pack was accumulated.
Bare ground was observed in both watersheds (50 – 70 percent of the landscape) with
snow accumulation areas measuring less than 15 cm (6 inches). Soil temperatures in
early February were approximately -5 degrees Celsius (22 degrees F). An early February
storm produced a rain on snow event. Flow was recorded in both watersheds and
evidence was observed which indicated the majority of channel flow originated as
overland flow. Sediment movement was observed on the hill slopes and in the channels.
Sediment deposits had to be removed from both flumes. These two different
observations help to illustrate the high variability within these systems and the difficulty
in connecting channel flow data to treatment effects, especially during the first two years
following treatment.
FUTURE RESEARCH THOUGHTS
Over the course of a long term study, there is always the occasional thought, what
should we have done differently if we knew then, what we know now. Fisher (2004)
pointed to the need to keep it simple. Collection, organization and analysis has been
overwhelming when dealing with 4 to 14 years of eight different sets of data.
89
When this study was first established, I believe that no one thought it would last
this long. Now, the commitment has been made to continue monitoring for the
foreseeable future. In addition, other similar sites are being explored throughout eastern
Oregon to further our understanding of semi-arid land function and in particular, the
hydrologic impacts of western juniper control on our landscapes. It is my hope that these
thoughts will provide some guidance for the initiation of those efforts.
Long term studies require the commitment of time, resources and personal
investment. Time and resources may be the easiest to schedule and plan for, personal
investment ebbs and flows as times change. First and foremost, when initiating this type
of study, maintaining relationships and communication with those directly involved and
as time permits, with those who will become involved as time goes forward. In the 12
years that transpired before treatment was imposed in Mays, leadership roles in the BLM
changed, leadership in ranching operations changed and leadership in the research roles
changed. Maintaining the focus of the study, its importance in developing our
understanding of watershed function while staying relevant to the current needs and
anticipate future questions and opportunities was always a challenge. Regularly
scheduled field visits, emails and written reports are useful in keeping everyone engaged
and the project moving forward.
The adoption of technology has made the job of data collection and storage
manageable. During the early years of the study, periods of time (as large as months)
would go by and due to technical problems, events (ex. channel flow) would be missed.
With the installation of the satellite communication technology, sensors could now be
checked weekly with the simple ease of logging onto the internet and going to the
90
website to see that data were being collected. Sensors not working could be identified
and repairs scheduled with the supporting agencies. Even with this new found capability,
the necessity for regular field visits was never eliminated.
Satellite communication technology provided the ease of data acquisition in an
area where other types of communication would have been difficult (cell phone or radio).
Acquisition of data via satellite communication was never a problem. That link in the
flow of data (satellites being available) was always there.
Understanding data needs and adjusting the timing of collection is important.
Weather data were collected 24 times a day in each watershed. Other sensors collected
data 4 times a day. For many of the parameters (i.e., soil moisture, soil temperature and
spring flow, ground water wells) once a day collection would have provided enough data
to see the changes that occurred with time.
With the satellite technology, I would explore the use of newer sensors (analog in
nature) that would provide data collection and storage during the periods between
satellite acquisition. The precipitation sensor is an example of a sensor that provides this
type of flexibility and detail in real time system monitoring. Precipitation data included
rain events between periods of data acquisition but all other weather parameters reflected
what was happening in the watersheds and the very moment the satellite acquired the
station. The sensor sensitivity was acceptable, however, we probably missed periods of
high wind gusts, by a degree or two missed the daily high and low temperatures, and
channel flow measured by the system. For channel flow, we had redundant systems and
the Campbell Scientific data logger installed at the initiation of the study provided more
frequent measurements (every 10 minutes).
91
As new ideas are generated and funding opportunities come along, one should not
be afraid to add monitoring to the system. The paired watershed project started with
precipitation and channel flow being monitored on a daily basis. Our understanding of
paired-watershed studies conducted primarily in forested systems provided this early
framework. As time progressed and observations were made, the question was asked, “if
we didn’t see changes in channel flow, how would change in watershed function have
been noted”? Through observed seasonal springflow and seepage in both watersheds, a
review of studies which focused on changes in erosion rates and vegetation, and the
review of early ground water findings, it was determined that additional monitoring was
needed in order to understand changes in soil moisture, ground water availability and
springflow discharge. While it would have been desirable to have the same length of
time on these parameters prior to treatment as was available for channel flow, channel
morphology and hillslope erosion, the information we did have was useful in noting
change.
The value of this study will come as additional studies are initiated and replication
in the study is conducted. In order for the results of this study to have meaning across the
6 million acres now occupied by western juniper, additional studies will be needed in
different landscapes and different soil types in different regions of eastern Oregon. It is
my hope that those studies will take advantage of the lessons learned during the last 14
years of the Camp Creek Paired Watershed Study, keep what has worked, and develop
new strategies for furthering our understanding of western juniper’s impact on the
hydrologic function of the landscapes it now occupies.
92
CONCLUSIONS
Anecdotal observations of watershed restoration activities across the west suggest
that ground water responds quickly to improved watershed conditions; vegetative
response follows shortly after leading to a reduction in erosion. These responses have
been observed to occur within a matter of months or at least within a very few years.
Channel morphological changes have been the last to occur and it has been surmised to
take years, even decades, depending on site capabilities and extent of the channel
degradation at the time of the restoration activity. The Camp Creek Paired Watershed
Study was initiated to provide long term, statistical data collected systematically in order
to test the hypothesis created by these long term observations and experiences, that being:
Does the removal of western juniper change the hydrologic
function of a watershed?
The hydrology of a given watershed is a function of the precipitation amount,
intensity and seasonality, watershed geomorphic characteristics including topography and
soil characteristics, and vegetation composition and complexity (Bowns 1999). The
essence of watershed management has been described as the “soil’s ability to Capture,
Store and Safely Release” precipitation where it falls. Watershed management as a
concept reaches beyond just water management and incorporates the importance of all the
resources and how they are interconnected. Those resources include plant, animal, soil
and climate as expressed through precipitation, temperature, wind, solar radiation, etc.
93
Capture refers to the processes in which precipitation enters the soil profile. Soil
surface roughness provided by plants, litter, rock and micro-terrain encourage the process
of infiltration. Storage of that moisture relies on the soil profile to hold moisture for plant
growth or allow it to move through the profile in to the ground water. Safe release is the
process of soil moisture being used for plant growth or excess soil moisture moving
through the soil profile, recharging ground water and expressing itself as either flow from
springs or as contributing to streamflow through seepage from stream banks (Bedell and
Borman 1997).
Land managers have the opportunity to affect watershed function through the
management of vegetation. Land managers, landowners and public land management
agencies through the application of management plans, impact vegetation’s ability to
buffer precipitation inputs and aid in the watersheds ability to capture, store and safely
release water.
SOIL MOISTURE
Juniper removal had a positive effect on the end of season (lowest recorded
reading of the year) soil moisture readings. In Mays, end of season soil moisture
readings increased at all profile depths (.2, .45 and .76 m). A significant difference was
found in the increased end of season moisture reading for the deepest profile for Mays
since treatment when compared to the pretreatment year. Changes in Jensen soil
moisture over the study period were not significant.
94
GROUND WATER
Juniper removal had a positive effect on the number of days that water was
recorded in the wells located in the treatment watershed (Mays). In Mays watershed,
well 6 has not gone dry since the junipers were cut. The two years prior to treatment, all
wells in Mays were dry by late summer. In 2007, all wells in Jensen went dry by late
September. Prior to this, well 4 in Jensen had never gone dry since depth to ground water
measurements began in January, 2004. Since treatment, the average increase in number
of days water is recorded in each well of Mays is 41.4 days while in Jensen, the average
number of days water was recorded since treatment increased 11.3 days.
SPRING FLOW
Juniper removal had a positive effect on the amount of spring flow in Mays.
Since treatment in 2005, flow in late season has increased 225 percent. Late season flow
is critical because it is during this period of time (late July through early November) that
precipitation is limited and flow is at its minimum. The increase in spring flow was
statistically significant even in a year in which the study area received only 80 percent of
its long term average annual rainfall (2006-07). During this dry year, the spring in the
control watershed (Jensen) went dry for the first time since flow monitoring began in
2004.
CHANNEL FLOW
Channel flow in the two watersheds was ephemeral. These channels were
influent with flow often going subsurface along the channel length and most often,
95
contributing to groundwater as opposed to groundwater contributing to channel flow.
Flow in these channels remained unpredictable and associated with snow melt and severe
summer thunderstorm activity. Channel flow, its presence or absence, responded very
quickly to its surrounding environment. If soils are dry or frozen, precipitation results in
flashy flows. If soils were moist and covered in snow, flow data from Mays indicated
that early season (spring time) flow was constant and its source was from bank seepage.
VEGETATION
At the end of the two year post treatment monitoring period, significant changes
in vegetative cover attributed to the treatment were increased canopy cover for perennial
forbs, annual forbs and an increased basal cover for annual grasses. Decreases in bare
soil and increases in shrub cover were measured but were not statistically significant.
Perennial grasses showed a slight decrease in basal cover which could be attributed to
observer error. Field observations in 2007 noted a large number of reproductive culms
per plant for all perennial grass species in the treated watershed while observations in the
control watershed noted very few reproductive culms per perennial grass plant. In 2008,
field observations and monitoring should begin to show an increase in plant density as a
result of increased seed production. Greenness of the grass plants was also noted.
Greenness (amount of green leaves within the plant) was greater in Mays when compared
to Jensen. Most of the plants in Jensen had gone dormant by late July, whereas, grass
plants in Mays appeared to still be photosynthesizing. Precipitation in the study area
during this monitoring period was 80 percent of the long term average.
96
HILLSLOPE EROSION
Hillslope erosion did not show any statistical changes as a result of treatment.
Precipitation events that would have caused significant soil erosion were absent during
the post treatment monitoring. As noted earlier, however, there was a rain on frozen
ground event the winter of 2007 but due to the frozen ground, minimal (not measurable)
hillslope soil movement was observed with that event. Summer thunderstorms in the
summer of 2006 generated observable soil movement in Jensen (control watershed) but
the impact of this storm was localized in only a part of that watershed and statistically
was not significant.
In Mays, hillslope erosion resulting in deposition of sediments into the channel
had been dampened due to the woody debris placed near the channel as a result of the
extraction activity which removed the juniper bole wood from the site. Woody debris
was observed to be acting as a filter, slowing any overland flow that occurred, thus
allowing the sediments to fall out before the flow entered the channel.
CHANNEL MORPHOLOGY
Channel morphology (the shape of the channel) was unique to each of the two
watersheds at the time the study was initiated. The channel in Jensen can be described as
being generally shallow, with less steep sides and a wider channel bottom. The channel in
Jensen could be characterized as U-shaped with the channel bottom controlled by rock.
The channel depth is rarely more than 3 feet deep. This channel also appeared to be
influenced more by side-hill erosion processes as demonstrated by channel cross-sections
being completely silted in due to side-hill sediment movement. The channel in Mays can
97
be characterized as being deeper and V-shaped with multiple head cuts found along its
length. This channel bottom was not controlled by rock.
The channel morphology data showed no statistical differences associated with
the treatment. Over the course of the monitoring period, channel data from Mays tended
to depict periodic but extreme soil movement events with intermixed periods of little to
no soil movement. Alternatively, the channel data from Jensen showed more consistent
soil movement with fewer extremes. This variation may be a product of the two channels
being at different evolutionary or successional stages relative to each other and thus
indicating that channel recovery would be different for each watershed.
By acknowledging these inherent differences within each watershed, we may
expect potentially different channel forming processes. Long term monitoring questions
yet to be answered include how ephemeral channel processes work following vegetation
treatments, and how recovery periods associated with vegetative treatments may differ
based on channel morphology, associated vegetative conditions and hydrologic
processes.
Hibbert (1983) summarized that arid and semiarid rangelands are poor water
producers because potential evapotranspiration is greater than precipitation. Hibbert
(1979) set a minimum threshold of 4500 mm (18 inches) of precipitation (4000 mm or 16
inches in cooler climates) as the point at which increases in water yield from perennial
stream systems could be measured as a result of vegetation manipulation. This study
provides evidence that these generalizations may be too conservative. The Camp Creek
Paired Watershed study has illustrated that in an ephemeral system, removing juniper for
the purpose of increasing water yield may be obtainable at a much lower precipitation
98
threshold. Seasonality of precipitation, with the majority of the annual precipitation
falling during the dormant season (October – April) may have been an important variable
in the measured water yield response occurring at the lower precipitation amount. As a
result of juniper removal, water yield increases were found in the amount of late season,
deep soil moisture, the amount of late season spring flow and number of days of available
near surface, ground water.
99
LITERATURE CITED
Alila, Y., M. Weiler, K. Green, D. Gluns and G. Jost. Not dated. Clearing the water:
Elucidating the relationship between equivalent cut area and watershed
hydrology: Moving beyond paired watershed studies. Accessed June, 2007.
www.forestry.ubc.ca/cottoncreek/files/extension_pairedws.pdf.
Anderson, W.E., Borman, M.M. and W. C. Krueger. 1998. The Ecological Provinces of
Oregon. SR 990, Oregon Agricultural Experiment Station. 138 p.
Automata Inc. 2005. Personnel communication.
Automata Service Manual. 2005. Sensors. In: Automata, Inc. Data Lynx Data
Management Tools. 75 p.
Azuma, D.L., B. A. Hiserote and P. A. Dunham. 2005. The Western juniper resource of
eastern Oregon. Resource Bulletin PNW-RB-249. USDA Forest Service Pacific
Northwest Research Station. 18 p.
Baker Jr., M.B. 1984. Changes in streamflow in an herbicide-treated pinyon-juniper
watershed in Arizona. Water Resources Research. 20(11):1639-1642.
Bates, J.D., R. F. Miller and T. Svejar. 2005. Long-term successional trends following
western juniper cutting. J. Rangeland Ecology and Management. 53(5):553-541.
Bedell, T.E. and M.M. Borman (editors). 1997 revised. Watershed Management Guide
for the Interior Northwest. Oregon State University Extension Service, EM-8436.
84 p.
Bowns, J.E. 1999. Ecology and management of pinyon-juniper communities within the
interior west: Overview of the “resource values session” of the symposium. IN:
Monson, S.B. and R. Stevens (compilers). Proceedings: Ecology and Management
of Pinyon-Juniper Communities within the Interior West. USDA Forest Service
Proceedings RMRS-P-9. Rocky Mtn. Research Station. Pgs. 157-163.
Bosch, J.M. and J.D. Hewlett. 1982. A review of catchment experiments to determine
the effect of vegetation changes on water yield and evapotranspiration. Journal of
Hydrology 55:2-23.
Buckhouse, J.C. and R. E. Gaither. 1982. Potential sediment production within
vegetative communities in Oregon’s Blue Mountains. J. Soil and Water
Conservation. 37(2):120-122.
100
Dodson, Elizabeth and T. Deboodt. (2007). Western Juniper Harvesting Methods for
Watershed Restoration and Wood Utilization: A Review. 60th Annual Meeting,
Society for Range Management. Reno, NV. Abstract no. 126.
Eddleman, L.E., P.M. Miller, R.F. Miller and P.L. Dysart. 1994. Western juniper
woodlands (of the pacific northwest). Interior Columbia Basin Ecosystem
Management Project (ICEBMP). Science Assessment. 132 p.
Ffolliott, P.F. and K.N. Brooks. 1988. Opportunities for enhancing water yield, quality,
and distribution in the Mountain West. IN: Schmidt, W.C., compiler. Proceedings
– future forests of the Mountain West: A stand culture symposium. USDA Forest
Service, Gen. Tech. Rep. INT-243 pp.55-60.
Fisher, M.P. 2004. Analysis of hydrology and erosion in small, paired watersheds in a
juniper-sagebrush area of central Oregon. PhD. Dissertation. Oregon State Univ.
Corvallis, OR. 220 p.
Fitzgerald, S.A. 1997. Forests and watershed management in eastern Oregon. IN: Bedell,
T.E. and M.M. Borman (editors). Watershed Management Guide for the Interior
Northwest. Oregon State University Extension Service. EM-8436. pgs. 55-61.
Gedney, D.R., D.L. Azuma, C.L. Bolsinger and N. McKay. 1999. Western juniper in
eastern Oregon. Gen. Tech. Rep. PNW-GTR-464. Portland, OR: USDA Forest
Service Pacific Northwest Research Station. 53 p.
Hatfield, Dr. P. 2001. Personal communication.
Hewlett, J.D., H.W. Lull and K.G. Reinhart. 1969. In Defense of Experimental
Watersheds. Water Resources Research. 5(1):306-316.
Hibbert, A.R. 1979. Managing vegetation to increase flow in the Colorado River Basin.
Gen. Tech. Rep. RM-66. USDA Forests Service Rocky Mountain Forest and
Range Experiment Station. 27 p.
Hibbert, A.R. 1983. Water yield improvement potential by vegetation management on
western rangelands. Water Resources Bulletin 19(3):375-381.
Huff, D.D., B. Hargrove, M.L. Tharp and R. Graham. 2000. Managing forests for water
yield; the importance of scale. Journal of Forestry: 15-19.
Ice, George G., and J.D. Stednick. 2004. Forest Watershed Research in the United States.
Forest History Today. Spring/Fall 2004: 16 – 26.
Jeppesen, D.J. 1978. Competitive moisture consumption by the western juniper
(Juniperus occidentalis). IN: Martin, R.E., J.E. Dealy and D.L. Caraher (editors).
Proceedings of the Western Juniper Ecology and Management Workshop
101
Johnsen, T.N., Jr. 1980. Picloram in water and soil from semiarid pinyon-juniper
watershed. Journal Environmental Quality. 9: 601-605.
Keppeler, E.T. 1998. The summer flow and water yield response to timber harvest.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-168. p. 35-43.
Larsen, R.E. 1993. Interception and water holding capacity of western juniper. PhD.
Dissertation. Oregon State Univ. Corvallis, OR. 172 p.
McCarthy III, F.J. and J.P. Dobrowolski. 1999. Hydrogeology and spring occurrence of
a disturbed juniper woodland in Rush Valley, Utah. IN: Monson, S.B. and R.
Stevens (compilers). Proceedings: Ecology and Management of Pinyon-Juniper
Communities within the Interior West. USDA Forest Service Proceedings RMRSP-9. Rocky Mtn. Research Station. Pgs. 194-199.
Megahan, W.F. and J. Hornbeck. 2000. Lessons learned in watershed management: A
retrospective view. USDA Forest Service Proceedings RMRS-P-13.2000. p.
177-188.
Miller, P.F. 1990. Physiological ecology of western juniper (Juniperus occidentalis
Hook. Subsp. occidentalis). PhD. Dissertation. Oregon State Univ. Corvallis, OR.
274 p.
Miller, R.F., J.D. Bates, T. Svejcar, F.B. Pierson and L.E. Eddleman. 2005. Biology,
ecology and management of western juniper. Technical Bulletin 152. Oregon
State University Agricultural Experiment Station. 77 p.
Monteith, S.S., J.M. Buttle, P.W. Hazlett, F.D. Beall, R.G. Semkin and D.S. Jeffries.
2006. Paired-basin comparison of hydrological response in harvested and
undisturbed hardwood forests during snowmelt in central Ontario: I. Streamflow,
groundwater and flowpath behaviour. Hydrological Processes 20:1095-1116.
Natural Resources Conservation Service. 2007. Crook County Soil Survey (draft).
Oregon Department of Water Resources. Deschutes Mitigation Program. Oregon
Administrative Rules, OAR 690-505 and 690-521.
Oregon Climate Service. Barnes Station 30 year Weather Data. Monthly Means and
Extremes, 1971 – 2000. www.ocs.oregonstate.edu/index.html
Roundy, B.A. and J.L. Vernon. 1999. Watershed values and conditions associated with
pinyon –juniper communities. IN: Proceedings: Ecology and Management of
Pinyon-Juniper Communities Within the Interior West. USDA Forest Service
Proceedings RMRS-9-9. p172-187.
102
Shepperd, W.D., C.E. Troendle and C.B. Edminster. 1992. Linking water and growth
and yield models to evaluate management alternatives in subalpine ecosystems.
IN: Getting to the Future through Silviculture. Gen. Tech. Rep. INT-291. USDA
Forest Service, Intermountain Research Station. Pg 42-48.
Sturges, D.L. 1994. High-elevation watershed response to sagebrush control. USDA
Forest Service Research Paper RM-318. Rocky Mountain Forest and Range
Experiment Station. 18 p.
Swank, G.W. 1969. Water yield improvement potentials on national forest lands tributary
to Ochoco Reservoir. USDA Forest Service Pacific Northwest Region. 52 p.
Thomas, L. 1995. Personal communication, soil classification. Bureau of Land
Management, Prineville District, Prineville, Oregon. July, 1995.
United States Army Corp of Engineers. 1980. Chapter 3 (Uplift and leakage). IN:
Engineering and Design – Instrumentation for Concrete Structures. Publication #
EM1110-2-4300. pg 3-11.
United States Department of Agriculture. 1966. Soil Survey, Prineville Area, Oregon.
Soil Conservation Service in cooperation with Oregon Agricultural Experiment
Station. Series 1955, No. 12. 89 p.
Wilcox, B.P. 1994. Runoff and erosion in intercanopy zones of pinyon-juniper
woodlands. Journal of Range Management. 47:285-295.
Wood, J.C., M.K. Wood and J.M. Tromble. 1987. Important factors influencing water
infiltration and sediment production on arid lands in New Mexico. Journal Arid
Envir. 12:111-118.
103
APPENDICES
104
APPENDIX A.
NRCS Soil Classifications for Mays and Jensen Watersheds, Draft 2008.
105
Map Unit Description (OR)
Crook County Area, Oregon
DRAFT - SUBJECT TO
089 - Luckycreek loam, 1 to 15 percent slopes
Mean annual precipitation: 12 to 16 inches
Mean annual temperature: 39 to 46 degrees F
Frost-free period: 40 to 70 days
Farmland class: Prime farmland if irrigated
Luckycreek and similar soils
Extent: about 85 percent of the unit
Landform(s): alluvial benches
fan terraces
small drainageways
stream terraces
Slope gradient: 1 to 12 percent
Parent material: fine-loamy alluvium from volcanic rock with
ash in the surface.
Restrictive feature(s): none
Seasonal high water table: greater than 60 inches
Flooding frequency: frequent
Ponding frequency: none
Representative soil profile:
A1
A2
2Bt1
2Bt2
2Bk1
2Bk2
-------
0
2
8
23
38
52
to
to
to
to
to
to
2 in
8 in
23 in
38 in
52 in
60 in
Texture
ashy loam
ashy loam
clay loam
gravelly clay loam
clay loam
very gravelly clay loam
Soil loss tolerance (T factor): 5
Wind erodibility group (WEG): 5
Wind erodibility index (WEI): 56
Land capability subclass, nonLand capability subclass, irrigated:
Drainage class: well drained
Hydric soil class: no
Hydrologic
C
Permeability
moderately rapid
moderately rapid
moderately slow
moderately slow
moderately slow
moderately slow
Available
Water Capacity
pH
Kw
Kf
0.3 to 0.4 in
0.6 to 1.2 in
1.5 to 3.1 in
1.5 to 3.2 in
1.4 to 2.9 in
0.8 to 1.7 in
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
7.4 to 7.8
7.9 to 8.4
.15
.20
.24
.20
.28
.10
.20
.28
.32
.32
.37
.37
Ecological Site / Plant Association: SR MOUNTAIN SWALE 12-16 PZ (R010XC017OR)
Survey Area Version: 0
Survey Area Version Date: 04/18/2007
4e
106
Map Unit Description (OR)
Crook County Area, Oregon
DRAFT - SUBJECT TO
245 - Anatone-Ateron-Polkbutte complex, 12 to 30 percent north slopes
Mean annual precipitation: 12 to 16 inches
Mean annual temperature: 39 to 46 degrees F
Frost-free period: 40 to 70 days
Farmland class: Not prime farmland
Anatone and similar soils
Extent: about 40 percent of the unit
Landform(s): hillslopes
mesas
mountain slopes
ridges
Slope gradient: 12 to 30 percent
Parent material: colluvium and residuum derived from
volcanic rock.
Restrictive feature(s): lithic bedrock at 10 to 20 inches
Seasonal high water table: greater than 60 inches
Flooding frequency: none
Ponding frequency: none
Representative soil profile:
A
BA
Bw1
Bw2
R
------
0
3
7
11
15
to
to
to
to
to
3 in
7 in
11 in
15 in
19 in
Texture
extremely cobbly ashy loam
very gravelly loam
very cobbly clay loam
extremely cobbly clay loam
unweathered bedrock
Soil loss tolerance (T factor): 1
Wind erodibility group (WEG): 8
Wind erodibility index (WEI): 0
Land capability subclass, nonLand capability subclass, irrigated:
Drainage class: well drained
Hydric soil class: no
Hydrologic
D
Permeability
moderately rapid
moderate
moderate
moderate
----
7s
Available
Water Capacity
pH
Kw
Kf
0.1 to 0.2 in
0.2 to 0.5 in
0.3 to 0.6 in
0.2 to 0.3 in
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
.02
.10
.05
.02
.28
.37
.28
.28
Ecological Site / Plant Association: JD MOUNTAIN CLAYPAN 12-16 PZ (R010XB080OR)
Ateron and similar soils
Extent: about 35 percent of the unit
Landform(s): south-facing mountain slopes
tops and south-facing hillslopes
Slope gradient: 12 to 30 percent
Parent material: residuum/colluvium from volcanic rock.
Restrictive feature(s): lithic bedrock at 12 to 20 inches
Seasonal high water table: greater than 60 inches
Flooding frequency: none
Ponding frequency: none
Soil loss tolerance (T factor): 1
Wind erodibility group (WEG): 8
Wind erodibility index (WEI): 0
Land capability subclass, nonLand capability subclass, irrigated:
Drainage class: well drained
Hydric soil class: no
Hydrologic
D
Survey Area Version: 0
Survey Area Version Date: 04/18/2007
7s
107
Map Unit Description (OR)
Crook County Area, Oregon
DRAFT - SUBJECT TO
Representative soil profile:
A
Bt1
Bt2
R
-----
0
5
12
18
to
to
to
to
5 in
12 in
18 in
28 in
Texture
extremely stony ashy loam
very cobbly clay loam
extremely stony clay
unweathered bedrock
Permeability
moderate
moderately slow
slow
----
Available
Water Capacity
pH
Kw
Kf
0.3 to 0.4 in
0.5 to 0.9 in
0.3 to 0.9 in
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
.05
.05
.02
.32
.28
.24
Ecological Site / Plant Association: SR MOUNTAIN SHALLOW NORTH 12-16 PZ (R010XC075OR)
Polkbutte, north and similar soils
Extent: about 15 percent of the unit
Landform(s): north-facing hillslopes
north-facing mountain slopes
Slope gradient: 12 to 30 percent
Parent material: volcanic ash over volcanic rock
Restrictive feature(s): lithic bedrock at 40 to 60 inches
Seasonal high water table: greater than 60 inches
Flooding frequency: none
Ponding frequency: none
Representative soil profile:
A -C -R --
0 to 3 in
3 to 42 in
42 to 52 in
Texture
ashy loamy sand
cobbly ashy loamy sand
unweathered bedrock
Soil loss tolerance (T factor): 5
Wind erodibility group (WEG): 7
Wind erodibility index (WEI): 180
Land capability subclass, nonLand capability subclass, irrigated:
Drainage class: well drained
Hydric soil class: no
Hydrologic
A
Permeability
rapid
rapid
----
Available
Water Capacity
pH
Kw
Kf
0.2 to 0.3 in
1.9 to 6.2 in
6.6 to 7.3
6.6 to 7.3
.10
.10
.17
.15
Ecological Site / Plant Association: JD ASHY DEEP NORTH 12-16 PZ (R010XB088OR)
Survey Area Version: 0
Survey Area Version Date: 04/18/2007
6e
108
Map Unit Description (OR)
Crook County Area, Oregon
DRAFT - SUBJECT TO
246 - Anatone-Ateron-Polkbutte complex, 30 to 65 percent north slopes
Mean annual precipitation: 12 to 16 inches
Mean annual temperature: 39 to 46 degrees F
Frost-free period: 40 to 70 days
Farmland class: Not prime farmland
Anatone, north and similar soils
Extent: about 45 percent of the unit
Landform(s): north-facing hillslopes
north-facing mountain slopes
ridges
Slope gradient: 30 to 65 percent
Parent material: colluvium and residuum derived from
volcanic rock.
Restrictive feature(s): lithic bedrock at 10 to 20 inches
Seasonal high water table: greater than 60 inches
Flooding frequency: none
Ponding frequency: none
Representative soil profile:
A
BA
Bw1
Bw2
R
------
0
3
7
11
15
to
to
to
to
to
3 in
7 in
11 in
15 in
19 in
Texture
extremely cobbly ashy loam
very gravelly loam
very cobbly clay loam
extremely cobbly clay loam
unweathered bedrock
Soil loss tolerance (T factor): 1
Wind erodibility group (WEG): 8
Wind erodibility index (WEI): 0
Land capability subclass, nonLand capability subclass, irrigated:
Drainage class: well drained
Hydric soil class: no
Hydrologic
D
Permeability
moderately rapid
moderate
moderate
moderate
----
7s
Available
Water Capacity
pH
Kw
Kf
0.1 to 0.2 in
0.2 to 0.5 in
0.3 to 0.6 in
0.2 to 0.3 in
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
.02
.10
.05
.02
.28
.37
.28
.28
Ecological Site / Plant Association: JD MOUNTAIN CLAYPAN 12-16 PZ (R010XB080OR)
Ateron, north and similar soils
Extent: about 35 percent of the unit
Landform(s): north-facing hillslopes
north-facing mountain slopes
north-facing ridges
Slope gradient: 30 to 65 percent
Parent material: residuum/colluvium from volcanic rock.
Restrictive feature(s): lithic bedrock at 12 to 20 inches
Seasonal high water table: greater than 60 inches
Flooding frequency: none
Ponding frequency: none
Soil loss tolerance (T factor): 1
Wind erodibility group (WEG): 8
Wind erodibility index (WEI): 0
Land capability subclass, nonLand capability subclass, irrigated:
Drainage class: well drained
Hydric soil class: no
Hydrologic
D
Survey Area Version: 0
Survey Area Version Date: 04/18/2007
7s
109
Map Unit Description (OR)
Crook County Area, Oregon
DRAFT - SUBJECT TO
Representative soil profile:
A
Bt1
Bt2
R
-----
0
5
12
18
to
to
to
to
5 in
12 in
18 in
28 in
Texture
extremely stony ashy loam
very cobbly clay loam
extremely stony clay
unweathered bedrock
Permeability
moderate
moderately slow
slow
----
Available
Water Capacity
pH
Kw
Kf
0.3 to 0.4 in
0.5 to 0.9 in
0.3 to 0.9 in
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
.05
.05
.02
.32
.28
.24
Ecological Site / Plant Association: SR MOUNTAIN SHALLOW NORTH 12-16 PZ (R010XC075OR)
Polkbutte, north and similar soils
Extent: about 15 percent of the unit
Landform(s): north-facing hillslopes
north-facing mountain slopes
Slope gradient: 30 to 65 percent
Parent material: volcanic ash over volcanic rock
Restrictive feature(s): lithic bedrock at 40 to 60 inches
Seasonal high water table: greater than 60 inches
Flooding frequency: none
Ponding frequency: none
Representative soil profile:
A -C -R --
0 to 3 in
3 to 42 in
42 to 52 in
Texture
ashy loamy sand
cobbly ashy loamy sand
unweathered bedrock
Soil loss tolerance (T factor): 5
Wind erodibility group (WEG): 7
Wind erodibility index (WEI): 180
Land capability subclass, nonLand capability subclass, irrigated:
Drainage class: well drained
Hydric soil class: no
Hydrologic
A
Permeability
rapid
rapid
----
Available
Water Capacity
pH
Kw
Kf
0.2 to 0.3 in
1.9 to 6.2 in
6.6 to 7.3
6.6 to 7.3
.10
.10
.17
.15
Ecological Site / Plant Association: JD ASHY DEEP NORTH 12-16 PZ (R010XB088OR)
Survey Area Version: 0
Survey Area Version Date: 04/18/2007
6e
110
Map Unit Description (OR)
Crook County Area, Oregon
DRAFT - SUBJECT TO
247 - Anatone-Ateron complex, 12 to 30 percent south slopes
Mean annual precipitation: 12 to 16 inches
Mean annual temperature: 39 to 46 degrees F
Frost-free period: 40 to 70 days
Farmland class: Not prime farmland
Anatone, south and similar soils
Extent: about 45 percent of the unit
Landform(s): hillslopes
mesas
mountain slopes
ridges
Slope gradient: 12 to 30 percent
Parent material: colluvium and residuum derived from
volcanic rock.
Restrictive feature(s): lithic bedrock at 10 to 20 inches
Seasonal high water table: greater than 60 inches
Flooding frequency: none
Ponding frequency: none
Representative soil profile:
A
BA
Bw1
Bw2
R
------
0
3
7
11
15
to
to
to
to
to
3 in
7 in
11 in
15 in
19 in
Texture
extremely cobbly ashy loam
very gravelly loam
very cobbly clay loam
extremely cobbly clay loam
unweathered bedrock
Soil loss tolerance (T factor): 1
Wind erodibility group (WEG): 8
Wind erodibility index (WEI): 0
Land capability subclass, nonLand capability subclass, irrigated:
Drainage class: well drained
Hydric soil class: no
Hydrologic
D
Permeability
moderately rapid
moderate
moderate
moderate
----
7s
Available
Water Capacity
pH
Kw
Kf
0.1 to 0.2 in
0.2 to 0.5 in
0.3 to 0.6 in
0.2 to 0.3 in
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
.02
.10
.05
.02
.28
.37
.28
.28
Ecological Site / Plant Association: JD MOUNTAIN CLAYPAN 12-16 PZ (R010XB080OR)
Ateron, south and similar soils
Extent: about 35 percent of the unit
Landform(s): south-facing mountain slopes
tops and south-facing hillslopes
Slope gradient: 12 to 30 percent
Parent material: residuum/colluvium from volcanic rock.
Restrictive feature(s): lithic bedrock at 12 to 20 inches
Seasonal high water table: greater than 60 inches
Flooding frequency: none
Ponding frequency: none
Soil loss tolerance (T factor): 1
Wind erodibility group (WEG): 8
Wind erodibility index (WEI): 0
Land capability subclass, nonLand capability subclass, irrigated:
Drainage class: well drained
Hydric soil class: no
Hydrologic
D
Survey Area Version: 0
Survey Area Version Date: 04/18/2007
7s
111
Map Unit Description (OR)
Crook County Area, Oregon
DRAFT - SUBJECT TO
Representative soil profile:
A
Bt1
Bt2
R
-----
0
5
12
18
to
to
to
to
5 in
12 in
18 in
28 in
Texture
extremely stony ashy loam
very cobbly clay loam
extremely stony clay
unweathered bedrock
Permeability
moderate
moderately slow
slow
----
Available
Water Capacity
pH
Kw
Kf
0.3 to 0.4 in
0.5 to 0.9 in
0.3 to 0.9 in
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
.05
.05
.02
.32
.28
.24
Ecological Site / Plant Association: SR MOUNTAIN SHALLOW SOUTH 12-16 PZ (R010XC054OR)
Area Version: 0
Survey Area Version Date: 04/18/2007
Survey
112
Map Unit Description (OR)
Crook County Area, Oregon
DRAFT - SUBJECT TO
248 - Anatone-Ateron complex, 30 to 65 percent south slopes
Mean annual precipitation: 12 to 16 inches
Mean annual temperature: 39 to 46 degrees F
Frost-free period: 40 to 70 days
Farmland class: Not prime farmland
Anatone, south and similar soils
Extent: about 45 percent of the unit
Landform(s): ridges
south-facing hillslopes
south-facing mountain slopes
Slope gradient: 30 to 65 percent
Parent material: colluvium and residuum derived from
volcanic rock.
Restrictive feature(s): lithic bedrock at 10 to 20 inches
Seasonal high water table: greater than 60 inches
Flooding frequency: none
Ponding frequency: none
Representative soil profile:
A
BA
Bw1
Bw2
R
------
0
3
7
11
15
to
to
to
to
to
3 in
7 in
11 in
15 in
19 in
Texture
extremely cobbly ashy loam
very gravelly loam
very cobbly clay loam
extremely cobbly clay loam
unweathered bedrock
Soil loss tolerance (T factor): 1
Wind erodibility group (WEG): 8
Wind erodibility index (WEI): 0
Land capability subclass, nonLand capability subclass, irrigated:
Drainage class: well drained
Hydric soil class: no
Hydrologic
D
Permeability
moderately rapid
moderate
moderate
moderate
----
7s
Available
Water Capacity
pH
Kw
Kf
0.1 to 0.2 in
0.2 to 0.5 in
0.3 to 0.6 in
0.2 to 0.3 in
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
.02
.10
.05
.02
.28
.37
.28
.28
Ecological Site / Plant Association: JD MOUNTAIN CLAYPAN 12-16 PZ (R010XB080OR)
Ateron, south and similar soils
Extent: about 40 percent of the unit
Landform(s): south-facing hillslopes
south-facing mountain slopes
Slope gradient: 30 to 65 percent
Parent material: residuum/colluvium from volcanic rock.
Restrictive feature(s): lithic bedrock at 12 to 20 inches
Seasonal high water table: greater than 60 inches
Flooding frequency: none
Ponding frequency: none
Soil loss tolerance (T factor): 1
Wind erodibility group (WEG): 8
Wind erodibility index (WEI): 0
Land capability subclass, nonLand capability subclass, irrigated:
Drainage class: well drained
Hydric soil class: no
Hydrologic
D
Survey Area Version: 0
Survey Area Version Date: 04/18/2007
7s
113
Map Unit Description (OR)
Crook County Area, Oregon
DRAFT - SUBJECT TO
Representative soil profile:
A
Bt1
Bt2
R
-----
0
5
12
18
to
to
to
to
5 in
12 in
18 in
28 in
Texture
extremely stony ashy loam
very cobbly clay loam
extremely stony clay
unweathered bedrock
Permeability
moderate
moderately slow
slow
----
Available
Water Capacity
pH
Kw
Kf
0.3 to 0.4 in
0.5 to 0.9 in
0.3 to 0.9 in
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
.05
.05
.02
.32
.28
.24
Ecological Site / Plant Association: SR MOUNTAIN SHALLOW SOUTH 12-16 PZ (R010XC054OR)
Survey Area Version: 0
Survey Area Version Date: 04/18/2007
114
Map Unit Description (OR)
Crook County Area, Oregon
DRAFT - SUBJECT TO
272 - Anatone-Bocker complex, 2 to 20 percent slopes
Mean annual precipitation: 12 to 16 inches
Mean annual temperature: 39 to 46 degrees F
Frost-free period: 40 to 70 days
Farmland class: Not prime farmland
Anatone and similar soils
Extent: about 45 percent of the unit
Landform(s): hillslopes
mesas
mountain slopes
ridges
Slope gradient: 2 to 20 percent
Parent material: colluvium and residuum derived from
volcanic rock.
Restrictive feature(s): lithic bedrock at 10 to 20 inches
Seasonal high water table: greater than 60 inches
Flooding frequency: none
Ponding frequency: none
Representative soil profile:
A
BA
Bw1
Bw2
R
------
0
3
7
11
15
to
to
to
to
to
3 in
7 in
11 in
15 in
19 in
Texture
extremely cobbly ashy loam
very gravelly loam
very cobbly clay loam
extremely cobbly clay loam
unweathered bedrock
Soil loss tolerance (T factor): 1
Wind erodibility group (WEG): 8
Wind erodibility index (WEI): 0
Land capability subclass, nonLand capability subclass, irrigated:
Drainage class: well drained
Hydric soil class: no
Hydrologic
D
Permeability
moderately rapid
moderate
moderate
moderate
----
7s
Available
Water Capacity
pH
Kw
Kf
0.1 to 0.2 in
0.2 to 0.5 in
0.3 to 0.6 in
0.2 to 0.3 in
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
.02
.10
.05
.02
.28
.37
.28
.28
Ecological Site / Plant Association: JD MOUNTAIN CLAYPAN 12-16 PZ (R010XB080OR)
Bocker and similar soils
Extent: about 40 percent of the unit
Landform(s): hillslopes
mesas
mountain slopes
ridges
Slope gradient: 2 to 20 percent
Parent material: colluvium/residuum from volcanic rock
Restrictive feature(s): lithic bedrock at 4 to 10 inches
Seasonal high water table: greater than 60 inches
Flooding frequency: none
Ponding frequency: none
Soil loss tolerance (T factor): 1
Wind erodibility group (WEG): 8
Wind erodibility index (WEI): 0
Land capability subclass, nonLand capability subclass, irrigated:
Drainage class: well drained
Hydric soil class: no
Hydrologic
D
Survey Area Version: 0
Survey Area Version Date: 04/18/2007
7s
115
Map Unit Description (OR)
Crook County Area, Oregon
DRAFT - SUBJECT TO
Representative soil profile:
A
BA
Bw
R
-----
0
3
7
10
to
to
to
to
3 in
7 in
10 in
14 in
Texture
very gravelly loam
very gravelly loam
very gravelly loam
unweathered bedrock
Permeability
moderate
moderate
moderate
----
Available
Water Capacity
pH
Kw
Kf
0.2 to 0.4 in
0.2 to 0.5 in
0.1 to 0.4 in
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
.05
.10
.10
.28
.32
.28
Ecological Site / Plant Association: JD VERY SHALLOW 12-16 PZ (R010XB032OR)
Survey Area Version: 0
Survey Area Version Date: 04/18/2007
116
Map Unit Description (OR)
Crook County Area, Oregon
DRAFT - SUBJECT TO
273 - Anatone-Bocker complex, 20 to 50 percent south slopes
Mean annual precipitation: 12 to 16 inches
Mean annual temperature: 39 to 46 degrees F
Frost-free period: 40 to 70 days
Farmland class: Not prime farmland
Anatone, south and similar soils
Extent: about 50 percent of the unit
Landform(s): hillslopes
mesas
mountain slopes
ridges
Slope gradient: 20 to 50 percent
Parent material: colluvium and residuum derived from
volcanic rock.
Restrictive feature(s): lithic bedrock at 10 to 20 inches
Seasonal high water table: greater than 60 inches
Flooding frequency: none
Ponding frequency: none
Representative soil profile:
A
BA
Bw1
Bw2
R
------
0
3
7
11
15
to
to
to
to
to
3 in
7 in
11 in
15 in
19 in
Texture
extremely cobbly ashy loam
very gravelly loam
very cobbly clay loam
extremely cobbly clay loam
unweathered bedrock
Soil loss tolerance (T factor): 1
Wind erodibility group (WEG): 8
Wind erodibility index (WEI): 0
Land capability subclass, nonLand capability subclass, irrigated:
Drainage class: well drained
Hydric soil class: no
Hydrologic
D
Permeability
moderately rapid
moderate
moderate
moderate
----
7s
Available
Water Capacity
pH
Kw
Kf
0.1 to 0.2 in
0.2 to 0.5 in
0.3 to 0.6 in
0.2 to 0.3 in
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
.02
.10
.05
.02
.28
.37
.28
.28
Ecological Site / Plant Association: JD MOUNTAIN CLAYPAN 12-16 PZ (R010XB080OR)
Bocker, south and similar soils
Extent: about 40 percent of the unit
Landform(s): hillslopes
mesas
mountain slopes
ridges
Slope gradient: 20 to 50 percent
Parent material: colluvium/residuum from volcanic rock
Restrictive feature(s): lithic bedrock at 4 to 10 inches
Seasonal high water table: greater than 60 inches
Flooding frequency: none
Ponding frequency: none
Soil loss tolerance (T factor): 1
Wind erodibility group (WEG): 8
Wind erodibility index (WEI): 0
Land capability subclass, nonLand capability subclass, irrigated:
Drainage class: well drained
Hydric soil class: no
Hydrologic
D
Survey Area Version: 0
Survey Area Version Date: 04/18/2007
7s
117
Map Unit Description (OR)
Crook County Area, Oregon
DRAFT - SUBJECT TO
Representative soil profile:
A
BA
Bw
R
-----
0
3
7
10
to
to
to
to
3 in
7 in
10 in
14 in
Texture
very gravelly loam
very gravelly loam
very gravelly loam
unweathered bedrock
Permeability
moderate
moderate
moderate
----
Available
Water Capacity
pH
Kw
Kf
0.2 to 0.4 in
0.2 to 0.5 in
0.1 to 0.4 in
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
.05
.10
.10
.28
.32
.28
Ecological Site / Plant Association: JD VERY SHALLOW 12-16 PZ (R010XB032OR)
Survey Area Version: 0
Survey Area Version Date: 04/18/2007
118
Map Unit Description (OR)
Crook County Area, Oregon
DRAFT - SUBJECT TO
391 - Ateron-Anatone-Erakatak complex, 0 to 50 percent slopes
Mean annual precipitation: 12 to 16 inches
Mean annual temperature: 39 to 46 degrees F
Frost-free period: 40 to 70 days
Farmland class: Not prime farmland
Ateron and similar soils
Extent: about 35 percent of the unit
Landform(s): hillslopes
mountain slopes
Slope gradient: 1 to 12 percent
Parent material: residuum/colluvium from volcanic rock.
Restrictive feature(s): lithic bedrock at 12 to 20 inches
Seasonal high water table: greater than 60 inches
Flooding frequency: none
Ponding frequency: none
Representative soil profile:
A
Bt1
Bt2
R
-----
0
5
12
18
to
to
to
to
5 in
12 in
18 in
28 in
Texture
extremely stony ashy loam
very cobbly clay loam
extremely stony clay
unweathered bedrock
Soil loss tolerance (T factor): 1
Wind erodibility group (WEG): 8
Wind erodibility index (WEI): 0
Land capability subclass, nonLand capability subclass, irrigated:
Drainage class: well drained
Hydric soil class: no
Hydrologic
D
Permeability
moderate
moderately slow
slow
----
7s
Available
Water Capacity
pH
Kw
Kf
0.3 to 0.4 in
0.5 to 0.9 in
0.3 to 0.9 in
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
.05
.05
.02
.32
.28
.24
Ecological Site / Plant Association: SR MOUNTAIN VERY SHALLOW 12-16 PZ (R010XC039OR)
Anatone and similar soils
Extent: about 25 percent of the unit
Landform(s): hillslopes
mountain slopes
ridges
Slope gradient: 1 to 12 percent
Parent material: colluvium and residuum derived from
volcanic rock.
Restrictive feature(s): lithic bedrock at 10 to 20 inches
Seasonal high water table: greater than 60 inches
Flooding frequency: none
Ponding frequency: none
Soil loss tolerance (T factor): 1
Wind erodibility group (WEG): 8
Wind erodibility index (WEI): 0
Land capability subclass, nonLand capability subclass, irrigated:
Drainage class: well drained
Hydric soil class: no
Hydrologic
D
Survey Area Version: 0
Survey Area Version Date: 04/18/2007
7s
119
Map Unit Description (OR)
Crook County Area, Oregon
DRAFT - SUBJECT TO
Representative soil profile:
A
BA
Bw1
Bw2
R
------
0
3
7
11
15
to
to
to
to
to
3 in
7 in
11 in
15 in
19 in
Texture
extremely cobbly ashy loam
very gravelly loam
very cobbly clay loam
extremely cobbly clay loam
unweathered bedrock
Permeability
moderately rapid
moderate
moderate
moderate
----
Available
Water Capacity
pH
Kw
Kf
0.1 to 0.2 in
0.2 to 0.5 in
0.3 to 0.6 in
0.2 to 0.3 in
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
.02
.10
.05
.02
.28
.37
.28
.28
Ecological Site / Plant Association: JD MOUNTAIN CLAYPAN 12-16 PZ (R010XB080OR)
Erakatak and similar soils
Extent: about 25 percent of the unit
Landform(s): hillslopes
mountain slopes
Slope gradient: 1 to 12 percent
Parent material: colluvium/residuum from volcanic rock with a
ash surface mantle
Restrictive feature(s): lithic bedrock at 20 to 40 inches
Seasonal high water table: greater than 60 inches
Flooding frequency: none
Ponding frequency: none
Representative soil profile:
A
2Bt1
2Bt2
2Bt3
2Bt4
R
-------
0
3
8
16
26
28
to
to
to
to
to
to
3 in
8 in
16 in
26 in
28 in
32 in
Texture
very gravelly ashy loam
very gravelly ashy clay loam
very cobbly clay
very cobbly clay
extremely cobbly silty clay
unweathered bedrock
Soil loss tolerance (T factor): 2
Wind erodibility group (WEG): 4
Wind erodibility index (WEI): 86
Land capability subclass, nonLand capability subclass, irrigated:
Drainage class: well drained
Hydric soil class: no
Hydrologic
C
Permeability
moderate
moderately slow
slow
slow
slow
----
Available
Water Capacity
pH
Kw
Kf
0.2 to 0.4 in
0.3 to 0.7 in
0.4 to 0.8 in
0.5 to 1.0 in
0.1 to 0.2 in
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
7.4 to 7.8
.05
.05
.05
.05
.02
.28
.32
.24
.20
.20
Ecological Site / Plant Association: JD MOUNTAIN CLAYPAN 12-16 PZ (R010XB080OR)
Survey Area Version: 0
Survey Area Version Date: 04/18/2007
4e
120
Map Unit Description (OR)
Crook County Area, Oregon
DRAFT - SUBJECT TO
430 - Gerow silt loam, drained, 1 to 5 percent slopes
Mean annual precipitation: 16 to 22 inches
Mean annual temperature: 39 to 45 degrees F
Frost-free period: 30 to 70 days
Farmland class:
Gerow, Drianed and similar soils
Extent: about 85 percent of the unit
Landform(s): incised stream terraces
Slope gradient: 1 to 5 percent
Parent material: alluvium from volcanic rock with volcanic ash
throughout.
Soil loss tolerance (T factor): 5
Wind erodibility group (WEG): 5
Wind erodibility index (WEI): 56
Land capability subclass, non-
6c
Land capability subclass, irrigated:
Restrictive feature(s): none
Drainage class: somewhat poorly drained
Seasonal high water table: approximately 52 inches
Hydric soil class: yes
Flooding frequency: rare
Hydrologic
C
Ponding frequency: none
Representative soil profile:
A
2AB
2Bwb1
2Bwb2
2Bwb3
2Bwb4
-------
0
6
10
15
25
36
to
to
to
to
to
to
6 in
10 in
15 in
25 in
36 in
60 in
Texture
ashy silt loam
ashy silty clay loam
ashy silty clay loam
ashy silty clay loam
silty clay
silty clay loam
Permeability
Available
Water Capacity
pH
Kw
Kf
1.2 to 1.4 in
0.8 to 0.9 in
1.1 to 1.2 in
2.1 to 2.4 in
1.5 to 2.2 in
3.4 to 5.0 in
5.6 to 6.0
5.6 to 6.0
5.6 to 7.8
5.6 to 7.8
6.1 to 8.4
6.1 to 8.4
.24
.32
.32
.32
.20
.32
.24
.32
.32
.32
.20
.32
moderate
moderately slow
moderately slow
moderately slow
moderately slow
moderately slow
Ecological Site / Plant Association: MOUNTAIN MEADOW (R010XY002OR)
Survey Area Version: 0
Survey Area Version Date: 04/18/2007
121
Map Unit Description (OR)
Crook County Area, Oregon
DRAFT - SUBJECT TO
P500 - Canest-Anatone complex, 0 to 10 percent slopes
Mean annual precipitation: 12 to 24 inches
Mean annual temperature: 39 to 46 degrees F
Frost-free period: 30 to 90 days
Farmland class: Not prime farmland
Canest and similar soils
Extent: about 70 percent of the unit
Landform(s): lava plains
plateaus
Slope gradient: 0 to 10 percent
Parent material: residuum weathered from basalt
Restrictive feature(s): lithic bedrock at 5 to 10 inches
Seasonal high water table: greater than 60 inches
Flooding frequency: none
Ponding frequency: none
Representative soil profile:
A -Bt -2R --
0 to 5 in
5 to 9 in
9 to 14 in
Texture
very cobbly clay loam
very cobbly clay
unweathered bedrock
Soil loss tolerance (T factor): 1
Wind erodibility group (WEG): 6
Wind erodibility index (WEI): 48
Land capability subclass, nonLand capability subclass, irrigated:
Drainage class: well drained
Hydric soil class: no
Hydrologic
C
Permeability
moderately slow
slow
----
7s
Available
Water Capacity
pH
Kw
Kf
0.4 to 0.7 in
0.2 to 0.4 in
6.6 to 7.3
6.6 to 7.3
.10
.10
.32
.28
Ecological Site / Plant Association: JD VERY SHALLOW 12-16 PZ (R010XB032OR)
stiff sagebrush/Sandberg bluegrass (SD9111)
Anatone and similar soils
Extent: about 15 percent of the unit
Landform(s): hillslopes
summits of mountain slopes
Slope gradient: 0 to 10 percent
Parent material: colluvium and residuum derived from basalt
or rhyolite.
Restrictive feature(s): lithic bedrock at 10 to 20 inches
Seasonal high water table: greater than 60 inches
Flooding frequency: none
Ponding frequency: none
Soil loss tolerance (T factor): 1
Wind erodibility group (WEG): 8
Wind erodibility index (WEI): 0
Land capability subclass, nonLand capability subclass, irrigated:
Drainage class: well drained
Hydric soil class: no
Hydrologic
D
Survey Area Version: 0
Survey Area Version Date: 04/18/2007
7s
122
Map Unit Description (OR)
Crook County Area, Oregon
DRAFT - SUBJECT TO
Representative soil profile:
A
BA
Bw1
Bw2
R
------
0
3
7
11
15
to
to
to
to
to
3 in
7 in
11 in
15 in
19 in
Texture
extremely cobbly ashy loam
very gravelly loam
very cobbly clay loam
extremely cobbly clay loam
unweathered bedrock
Permeability
moderately rapid
moderate
moderate
moderate
----
Available
Water Capacity
pH
Kw
Kf
0.1 to 0.2 in
0.2 to 0.5 in
0.3 to 0.6 in
0.2 to 0.3 in
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
6.6 to 7.3
.02
.10
.05
.02
.28
.37
.28
.28
Ecological Site / Plant Association: JD MOUNTAIN CLAYPAN 12-16 PZ (R010XB080OR)
Survey Area Version: 0
Survey Area Version Date: 04/18/2007
123
APPENDIX B.
Partial Watershed Plant Species List (Fisher 2004)
Perennial Grasses
Agropyron spicatum
Elymus glaucus var. glaucus
Festuca idahoensis
Koeleria cristata
Oryzopsis humenoides
Poa ampla
Poa sandbergii
Olypogon monspeliensis
Sitanion hystrix
Stipa comata
Stipa occidentalis
bearded bluebunch wheatgrass
blue wildrye
Idaho fescue
prairie junegrass
Indian ricegrass
big bluegrass
sandberg bluegrass
rabbitfoot grass
bottlebrush squirreltail
needle-and-thread
western needlegrass (Columbiana)
Sedge-Rush
Carex geyeri
elk sedge
Annual Grasses
Bromus tectorum
cheatgrass brome
Perennial Forbs
Achillea millefolium
Antennaria stenophylla
Arenaria frankinii
Balzamorhiza sagittata
Calacohortus macrocarpus
Chaenactis douglasii
Erigeron linearis
Gilia aggregate, var. aggregata
Geum trifolium
Lithopermum ruderale
Lupinus spp.
Penstemon eriantherus
Phacelia hastate
Salvia dorrii
Senecio integerrimus
common yarrow
pussy toes
sandwort
arrowleaf balsamroot
sagebrush mariposa
Douglas c. / false yarrow
lineleaf fleabane (yellow)
skyrocket scarlet gilia
prairie star
stoneseed
lupine
fuzzy tongue
silverleaf phacelia
greyball sage
western groundsel
124
Biennial Forbs
Tragopogan dubius
yellow salsify
Annual Forbs
Collinsia grandoflora
Collomia grandiflora
Cordylanthus ramosus
Erysimum occidentale
Lygodesmia juncea
Mimulus breweri
pagoda plant
bigflower gilia
bushy birdbeak
pale wallflower
rush skeletonweed
crimson monkeyflower
Shrubs and Trees
Amelanchier alnifolia
Artemisia ludoviciana
Artemisia tridentata var. tridentata
Artemisia tridentata var. wyomingensis
Artemisia tridentata var. vaseyana
Atriplex spinosa
Cercocarpus ledifolius
Chrysothamnus nauseosus
Chrysothamnum viscidiflorus
Holodiscus dumosus
Juniperous occidentalis
Pinus ponderosa
Purshia tridentata
Ribes cereum
Sambucus racemosa
Symphorycarpos albus
Tetradymia canescens
pacific serviceberry
prairie sage
basin big sagebrush
Wyoming big sagebrush
mountain big sagebrush
curl-leaf mountain mahogany
gray rabbitbrush
green rabbitbrush
little oceanspray
western juniper
ponderosa pine
bitterbrush
wax current
elderberry
snowberry
gray spineless horsebrush
125
APPENDIX C: Story of June 28 – 29, 2006 thunderstorm in Mays and Jensen.
On June 28th, 2006, a late-evening thunderstorm occurred over the paired watershed
study area. In one hour, precipitation in Jensen (red) weather station measured 190 mm
(.75 inches) and Mays weather station (blue) recorded 165mm (.65 inches) of rain. At the
end of the storm (15 hours), total precipitation received measured 289 mm (1.2 inches) in
Jensen and 279 mm (1.1 inches) in Mays.
Within 1 hour of the start of the storm, sensors in Jensen flume (red) recorded flow of .4
ft on the staff gauge. Flow in Mays (blue) was not detected. The continuous flow in
Jensen after the initiation of the storm reflects sediment deposits in the flume and not
actual flow. Flow lasted less than 6 hours.
126
Jensen flume 4 days after summer
thunderstorm, notice sediment deposits
in the floor of the flume.
Mays flume 4 days after the same storm.
Floor of flume has dusting of sediment
but floor is already dry.
Staff gauge in Jensen flume showed that
actual flow may have been as high as
22.86 cm (9 inches). The bottom of the
photo shows the height of mud in the
flume.
Staff gauge in Mays flume showed that
rain drop splatter scattered mud on the
walls of the flume but no flow line is
visible.
The sensor posed over the throat of the
flume is an ultrasonic sensor, measuring
the changes in distance between the
sensor and bottom of flume. Sediment
deposits are evident on the flume floor
and measure 10 cm (4 inches).
The ultrasonic sensor in Mays recorded
no changes in depth of flow. Sediments
on the flow of the flume are less than 31
mm (1/8 inch) deep.
127
Channel deposits in Jensen. This deposit
was laid across channel cross-section
number 4. The stake on the far side of
the channel was buried with deposits to
the top of the stake. Channel cross
section here is 338 cm (133 inches).
The channels of Mays showed no
evidence of large flows or sediment
deposits. This photo shows vegetation
and a small amount of channel flow just
above Mays’ flume. Flow disappears
before crossing through the fence.
Near Jensen cross-section number 22,
overland flow was evident. Overland
flow moved sediment and organic
material into the channel.
This road was used as a skid trail. Just
days before the storm, road condition
was soft, with several inches of loose
soil on top.
Channel deposits were numerous in
Jensen following the storm. These
deposits were place near Jensen spring.
One of two channels monitored in Mays.
Road used for skidding is on the right.
Evidence of channel flow was minimal.
128
APPENDIX D: Vegetative monitoring changes from 2005 and 2007
Jensen vegetation transect #1, 7/24/07
Jensen vegetation transect #1, 7/25/05
Jensen vegetation transect #2, 7/25/05
Jensen vegetation transect #2, 7/24/07
Jensen vegetation transect #3, 7/25/05
Jensen vegetation transect #3, 7/24/07
129
Jensen vegetation transect #4, 7/25/05
Jensen vegetation transect #4, 7/24/07
Jensen vegetation transect #5, 7/25/05
Jensen vegetation transect #6, 7/24/07
Jensen vegetation transect #6, 7/25/05
Jensen vegetation transect #6, 7/24/07
130
Jensen vegetative transect #7, 7/25/05
Jensen vegetative transect #7, 7/24/07
Jensen vegetative transect #8, 7/25/05
Jensen vegetative transect #8, 7/24/07
131
Mays vegetative transect #1, 7/26/05
Mays vegetative transect #1, 7/26/07
Mays vegetative transect #2, 7/26/05
Mays vegetative transect #2, 7/26/07
Mays vegetative transect #3, 7/26/05
Mays vegetative transect #3, 7/26/07
132
Mays vegetative transect #4, 7/26/05
Mays vegetative transect #4, 7/26/07
Mays vegetative transect #5, 7/26/05
Mays vegetative transect #5, 7/26/07
Mays vegetative transect #6, 7/26/05
Mays vegetative transect #6, 7/26/07
133
Mays vegetative transect #7, 7/26/05
Mays vegetative transect #7, 7/26/07
Mays vegetative transect #8, 7/26/05
Mays vegetative transect #8, 7/26/07
134
APPENDIX E. Vegetative responses, pictures from 7/25/07 – 7/28/07
Overview of Jensen Watershed
General view of the understory condition in Jensen. Grasses are mature and
mostly dormant. Slope is generally easterly. There is very little “greenness” left.
135
Jensen watershed, another view of understory, north slope.
Jensen watershed, a south slope, very few reproductive culms are evident.
136
Mays watershed overview, pre treatment (July, 2005)
Mays watershed overview, post treatment (July, 2007).
137
Mays watershed, south slope highlighting the flush of cheatgrass.
Mays watershed - Reproductive culms
on Junegrass and Big blue.
Mays Watershed - Plant release and
reproduction of basin wildrye.
138
Mays watershed – Reproductive culms on bluebunch wheatgrass and Indian
ricegrass. Also visible is the reproductive release of mountain big sagebrush.
Mays watershed - Increase in perennial forb cover was noted. Crepis in bloom.
139
Mays watershed – Lupine spp. in bloom in late July.
140
Appendix F. Soil moisture probe locations.
Watershed
Probe
Location
Site
Probe
Number
Mays
Top
Lower
11303
11323
11342
11403
11411
11442
11304
11324
11343
11404
11424
11443
11322
11341
11344
11422
11423
11444
Upper
Middle
Lower
Upper
Bottom
Lower
Upper
Jensen
Top
Lower
Upper
Middle
Lower
Upper
Bottom
Lower
Upper
11703
11722
11741
11803
11822
11842
11704
11724
11743
11804
11823
11843
11723
11742
11744
11824
11841
11844